Ristie 5

ISSN: 1646-9895
Revista Ibérica de Sistemas e Tecnologias de Informação

Iberian Journal of Information Systems and Technologies
J u l h o 1 6 • J u l y 1 6
©AISTI 2016 http://www.aisti.eu Nº E5

Revista Ibérica de Sistemas e Tecnologias de Informação
Iberian Journal of Information Systems and Technologies
Edição / Edition
Nº. E5, 07/2016
ISSN: 1646-9895
Indexação / Indexing
Academic Journals Database, CiteFactor, Dialnet, DOAJ, DOI, EBSCO, EI-Compendex,
GALE, IndexCopernicus, Index of Information Systems Journals, Latindex, ProQuest,
QUALIS, SCImago, SCOPUS, SIS, Ulrich’s.
Propriedade e Publicação / Ownership and Publication

AISTI – Associação Ibérica de Sistemas e Tecnologias de Informação
Rua Quinta do Roseiral 76, 4435-209 Rio Tinto, Portugal
E-mail: aistic@gmail.com
Web: http://www.risti.xyz
RISTI, N.º E5, 07/2016 i

Ficha Técnica
Director
Álvaro Rocha, Universidade de Coimbra, PT
Coordenadores da Edição / Issue Coordinators

Ning Zhang, Renmin University of China, Beijing, CN
Álvaro Rocha, Universidade de Coimbra, PT
Conselho Editorial / Editorial Board

Carlos Ferrás Sexto, Universidad de Santiago de Compostela, ES
Gonçalo Paiva Dias, Universidade de Aveiro, PT
Jose Antonio Calvo-Manzano Villalón, Universidad Politécnica de Madrid, ES
Luís Paulo Reis, Universidade do Minho, PT
Manuel Pérez Cota, Universidad de Vigo, ES
Ramiro Gonçalves, Universidade de Trás-os-Montes e Alto Douro, PT
Conselho Científico / Scientific Board

Adolfo Lozano-Tello, Universidad de Extremadura, ES
Adrián Hiebra Pardo, Universidad de Santiago de Compostela, ES
Adriano Pasqualotti, Universidade de Passo Fundo, BR
Alberto Fernández, Universidad Rey Juan Carlos, ES
Alberto Bugarín, Universidad de Santiago de Compostela, ES
Alejandro Medina, Universidad Politécnica de Chiapas, MX
Alejandro Rodríguez González, Universidad Politécnica de Madrid, ES
Alejandro Peña, Escuela de Ingeniería de Antioquia, CO
Alexandre L’Erario, Universidade Tecnológica Federal do Paraná, BR
Alma María Gómez-Rodríguez, Universidad de Vigo, ES
Álvaro E. Prieto, Universidad de Extremadura, ES
Ana Azevedo, Instituto Politécnico do Porto, PT
Ana Cristina Ramada Paiva, FEUP, Universidade do Porto, PT
Ana Isabel Veloso, Universidade de Aveiro, PT
Ana Maria Correia, ISEGI, Universidade Nova de Lisboa, PT
Anabela Mesquita, Instituto Politécnico do Porto, PT
ii RISTI, N.º E5, 07/2016

Angelica Caro, Universidad del Bío-Bío, CL
Ania Cravero, Universidad de La Frontera, CL
Antoni Lluís Mesquida Calafat, Universitat de les Illes Balears, ES
Antonia Mas Pichaco, Universitat de les Illes Balears, ES
António Coelho, FEUP, Universidade do Porto, PT
António Godinho, ISLA-Gaia, PT
Antonio Jesus Garcia Loureiro, Universidad de Santiago de Compostela, ES
António Pereira, Instituto Politécnico de Leiria, PT
Armando Mendes, Universidade dos Açores, PT
Arnaldo Martins, Universidade de Aveiro, PT
Arturo J. Méndez, Universidad de Vigo, ES
Baltasar García Pérez-Schofield, Universidad de Vigo, ES
Benjamim Fonseca, Universidade de Trás-os-Montes e Alto Douro, PT
Bráulio Alturas, ISCTE - Instituto Universitário de Lisboa, PT
Brenda L. Flores-Rios, Universidad Autónoma de Baja California, MX
Brígida Mónica Faria, ESTSP, Instituto Politécnico do Porto, PT
Carlos Costa, Universidade de Aveiro, PT
Carlos Rabadão, Instituto Politécnico de Leiria, PT
Carlos Carreto, Instituto Politécnico da Guarda, PT
Carlos Morais, Instituto Politécnico de Bragança, PT
Carlos Vaz de Carvalho, Instituto Politécnico do Porto, PT
Carmen Galvez, Universidad de Granada, ES
César Gonçalves, Universidade do Algarve, PT
Ciro Martins, Universidade de Aveiro, PT
Cristina Alcaraz, Universidad de Málaga, ES
Daniel Castro Silva, Universidade de Coimbra, PT
Daniel Polónia, Universidade de Aveiro, PT
Daniel Riesco, Universidad Nacional de San Luis, AR
David Fonseca, Universitat Ramon Llull, ES
David Ramos Valcarcel, Universidad de Vigo, ES
Dora Simões, Universidade de Aveiro, PT
Eduardo Sánchez Vila, Universidad de Santiago de Compostela, ES
Emiliano Reynares, CIDISI - UTN FRSF - CONICET, AR
Enric Mor, Universitat Oberta de Catalunya, ES
RISTI, N.º E5, 07/2016 iii

Eusébio Ferreira da Costa, Escola Superior de Tecnologias de Fafe, PT
Feliz Gouveia, Universidade Fernando Pessoa, PT
Fernando Bandeira, Universidade Fernando Pessoa, PT
Fernando Diaz, Universidad de Valladolid, ES
Fernando Moreira, Universidade Portucalense, PT
Francisco Restivo, Universidade Católica Portuguesa, PT
Gerardo Gonzalez Filgueira, Universidad da Coruña, ES
Gerardo Rodriguez, Universidad de Salamanca, ES
Germano Montejano, Universidad Nacional de San Luis, AR
Guilhermina Lobato Miranda, Universidade de Lisboa, PT
Hélder Zagalo, Universidade de Aveiro, PT
Hélia Guerra, Universidade dos Açores, PT
Henrique Gil, Instituto Politécnico de Castelo Branco, PT
Henrique Santos, Universidade do Minho, PT
Higino Ramos, Universidad de Salamanca, ES
Hugo Paredes, Universidade de Trás-os-Montes e Alto Douro, PT
Isabel Pedrosa, Instituto Politécnico de Coimbra, PT
Isaura Ribeiro, Universidade dos Açores, PT
Isidro Calvo, Universidad del País Vasco (UPV/EHU), ES
Ismael Etxeberria-Agiriano, Universidad del País Vasco (UPV/EHU), ES
Ivan Garcia, Universidad Tecnologica de la Mixteca, MX
Javier Garcia Tobio, CESGA-Centro de Supercomputacion de Galicia, ES
Jezreel Mejia, Centro de Investigación en Matemática (CIMAT), MX
João Pascual Faria, FEUP, Universidade do Porto, PT
João Paulo Costa, Universidade de Coimbra, PT
João Tavares, FEUP, Universidade do Porto, PT
Joaquim José Gonçalves, Instituto Politécnico do Cávado e do Ave, PT
Joaquim Madeira, Universidade de Aveiro, PT
Joaquim Reis, ISCTE - Instituto Universitário de Lisboa, PT
Jörg Thomaschewski, University of Applied Sciences OOW - Emden, DE
Jose Alfonso Aguilar, Universidad Autonoma de Sinaloa, MX
José Augusto Fabri, Universidade Tecnológica Federal do Paraná, BR
José Braga de Vasconcelos, Universidade Atlântica, PT
José Cascalho, Universidade dos Açores, PT
iv RISTI, N.º E5, 07/2016

José Felipe Cocón Juárez, Universidad Autónoma del Carmen, MX
Jose J. Pazos-Arias, Universidad de Vigo, ES
José Luís Silva, Universidade da Madeira, PT
José Paulo Lousado, Instituto Politécnico de Viseu, PT
José Luis Pestrana Brincones, Universidad de Málaga
José Luís Reis, ISMAI - Instituto Superior da Maia, PT
Jose M Molina, Universidad Carlos III de Madrid, ES
Jose Maria Zavala Perez, Eticas Research & Consulting, ES
José Martins, Universidade de Trás-os-Montes e Alto Douro, PT
Jose Maria de Fuentes, Universidad Carlos III de Madrid, ES
Jose R. R. Viqueira, Universidade de Santiago de Compostela, ES
José Silvestre Silva, Academia Militar, PT
Josep M. Marco-Simó, Universitat Oberta de Catalunya, ES
Juan D’Amato, PLADEMA-UNCPBA-CONICET, AR
Juan M. Santos Gago, Universidad de Vigo, ES
Juan Manuel Fernández-Luna, Universidad de Granada, ES
Juan-Manuel Lopez-Zafra, Universidad Complutense de Madrid, ES
Leonardo Bermon, Universidad Nacional de Colombia, CO
Leila Weitzel, Universidade Federal Fluminense, BR
Lilia Muñoz, Universidad Tecnológica de Panamá, PA
Luis Alvarez Sabucedo, Universidad de Vigo, ES
Luís Correia, Universidade de Lisboa, PT
Luis de Campos, Universidad de Granada, ES
Luis Enrique, Sicaman Nuevas Tecnologías S.L., ES
Luis Fernandez-Sanz, Universidad de Alcalá, ES
Luís Ferreira, Instituto Politécnico do Cávado e do Ave, PT
Luis Vilán-Crespo, Universidad de Vigo, ES
Luisa María Romero-Moreno, Universidad de Sevilla, ES
Luisa Miranda, Instituto Politécnico de Bragança, PT
Lus Sussy Bayona Ore, Universidad Nacional Mayor de San Marcos, PE
Magdalena Arcilla Cobián, Universidade Nacional de Educación a Distancia, ES
Manuel Jose Fernandez Iglesias, Universidad de Vigo, ES
Marcelo Mendonça Teixeira, Universidade Federal Rural de Pernambuco, BR
Marco Painho, ISEGI, Universidade Nova de Lisboa, PT
RISTI, N.º E5, 07/2016 v

Maria Hallo, Escuela Politécnica Nacional, EC
María J. Lado, Universidad de Vigo, ES
Maria João Castro, Instituto Politécnico do Porto, PT
Maria João Ferreira, Universidade Portucalense, PT
Maria João Gomes, Universidade do Minho, PT
Maria José Angélico, Instituto Politécnico do Porto, PT
Maria José Marcelino, Universidade de Coimbra, PT
Maria José Sousa, Universidade Europeia, PT
Marisol B. Correia, Universidade do Algarve, PT
Maristela Holanda, Universidade de Brasília, BR
Martín Llamas Nistal, Universidad de Vigo, ES
Matías García Rivera, Universidad de Vigo, ES
Mercedes Ruiz, Universidad de Cádiz, ES
Miguel A. Brito, Universidade do Minho, PT
Miguel Bugalho, Universidade Europeia, PT
Miguel Casquilho, IST, Universidade de Lisboa, PT
Mirna Ariadna Muñoz Mata, Centro de Investigación en Matemáticas (CIMAT), MX
Nelson Rocha, Universidade de Aveiro, PT
Nuno Lau, Universidade de Aveiro, PT
Nuno Ribeiro, Universidade Fernando Pessoa, PT
Orlando Belo, Universidade do Minho, PT
Oscar Mealha, Universidade de Aveiro, PT
Paula Peres, Instituto Politécnico do Porto
Paula Prata, Universidade da Beira Interior, PT
Paulo Martins, Universidade de Trás-os-Montes e Alto Douro, PT
Paulo Pinto, FCT, Universidade Nova de Lisboa, PT
Pedro Abreu, Universidade de Coimbra, PT
Pedro Miguel Moreira, Instituto Politécnico de Viana do Castelo, PT
Pedro Nogueira Ramos, ISCTE - Instituto Universitário de Lisboa, PT
Pedro Sánchez Palma, Universidad Politécnica de Cartagena, ES
Pedro Sanz Angulo, Universidad de Valladolid, ES
Pilar Mareca Lopez, Universidad Politécnica de Madrid, ES
Raul Laureano, ISCTE - Instituto Universitário de Lisboa, PT
Renata Spolon Lobato, UNESP - Universidade Estadual Paulista, BR
vi RISTI, N.º E5, 07/2016

Reinaldo Bianchi, Centro Universitário da FEI, BR
Rita Santos, Universidade de Aveiro, PT
Roberto Rodrígues Echeverría, Universidad de extremadura, ES
Rodolfo Miranda Barros, Universidade Estadual de Londrina, BR
Rogério Eduardo Garcia, Universidade Estadual Paulista, BR
Rubén González Crespo, Universidad Internacional de La Rioja, ES
Rui Cruz, IST, Universidade de Lisboa, PT
Rui José, Universidade do Minho, PT
Rui Pedro Marques, Universidade de Aveiro, PT
Santiago Gonzales Sánchez, Universidad Inca Garcilaso de la Vega, PE
Sergio Gálvez Rojas, Universidad de Málaga, ES
Sérgio Guerreiro, Universidade Lusófona de Humanidades e Tecnologias, PT
Silvia Fernandes, Universidade do Algarve, PT
Solange N Alves de Souza, Universidade de São Paulo, BR
Tomás San Feliu Gilabert, Universidad Politécnica de Madrid, ES
Valéria Farinazzo Martins, Universidade Presbiteriana Mackenzie, BR
Victor Hugo Medina Garcia, Universidad Distrital Francisco José de Caldas, CO
Vitor Carvalho, Instituto Politécnico do Cávado e do Ave, PT
Vitor Santos, ISEGI, Universidade Nova de Lisboa, PT
Wagner Tanaka Botelho, Universidade Federal do ABC, BR
Xose A. Vila, Universidad de Vigo, ES
RISTI, N.º E5, 07/2016 vii

Editorial
Information Technology Drives Innovation in Service
Ning Zhang*
* zhangninruc@163.com
Renmin University of China, Beijing, China

ix
Technological innovation presents a diversified development trend, with the integration

of information technology services and industries, cloud computing, big data and other
new models has been recognized by more and more users. Information technology
services companies face the pressure of transformation and upgrading, business model
change is underway.
Information technology services industry as an important part of the IT industry, it can
provide digital platform development, business data storage and processing, system
design and consulting services, and will promote productivity and economic growth.
Thus, the RISTI (Revista Iberica de Sistemas e Tecnologias de Informacao) presents
actual papers on main topics of journal specialization to address and focus new innovative
in information technologies. The Vol. 2016, No. E5, has selected 40 extended articles
presented at the panel discussion on Information Technology Innovation and Computer
Engineering in July 2015 in Beijing China, which aims to more deep insights about the
development of “Information Technology Drives Innovation in Service”.
We gratefully appreciate the authors for their scholarly contributions, and all the
included contents were anonymously reviewed by distinguished experts to maintain
academic excellence and integrity. We assure our readers that RISTI (Revista Iberica
de Sistemas e Tecnologias de Informacao) will serve as a key exchange platform for
publications in information science & technology field. The authors and readers can
count on the journal.
RISTI, N.º E5, 07/2016 ix

Índice / Index
EDITORIAL
Information Technology Drives Innovation in Service �� ix
Ning Zhang
ARTIGOS / ARTICLES
Research on Influence Factors of New Rural Sports Culture Development
Based on Data Mining Technology��1
Huasheng Huang, Xiaojun Wu, Ganchen Tao, Ying Zhang, Yuefeng Wang, Xiaodong Long
Influence of Corporate Equity Incentive and CSR on Firm Performance:
an Empirical Analysis based on Large-scale Data Mining��13
Ming Li, Zhenxiang Chen
Study on the Image Supporting System and Propagation Mechanism
of Rural Tourism Destination......................................................................................... 26
Li Long, Song Hui
Countermeasures and Support Environment for the Development of the
Non-governmental Economy in Suzhou, China..............................................................37
TianXueying, NiePanke
Effectiveness of College Counselors’ Work and Ideological
Education Based on Multimedia Platform..................................................................... 50
Pingqian Guo, Yafeng Zhang
The Influence of Strategic HRM on Enterprise Performance based on
Online Survey and Data Mining.......................................................................................61
Shuxia Wang
Asymmetric Effect of Energy prices on Energy Saving and
Emission Reduction based on Dynamic panel GMM Test.............................................. 71
Hua Peng
BIRCH Algorithm and Data Mining Application in Construction of Financial
Enterprise Management Team based on “Internet Plus” Background ......................... 84
Chenglin Xiao, Weili Xia
Urbanization Moderate scale Management of Land and Grain
Production based on Evolutionary Algorithm ................................................................95
Tao Zhai, Yadong Fan
E-course of College Badminton Curriculum Based on Multimedia Platform:
an Experiential Teaching Method..................................................................................106
Hong Zhang
x RISTI, N.º E5, 07/2016

Research on Badminton Footwork Training and Teaching Effect based on
Computer Aided Information Processing...................................................................... 115
Chao Chen
Study on the Cooperation of Hotels and Online Self-service Travel Business
based on Information Technology.................................................................................125
Xiaolong He, Xiaomei Zhang, Fangyuan Liu
Prediction and Simulation on the Consumers Purchase Intention of
Rice based on System Dynamics Modeling...................................................................136
Shengde Hu, Haoning Wu, Yuqiu Cai
Research on Evaluation Index System of Tourism Competitiveness
based on Tourism E-business Platform ........................................................................146
Wu Ling, Li Long
Research on Sports Tourism Resources Value and Promotion with Skating
and Skiing based on Data Mining Method....................................................................158
Xianfeng Zha, Xiaomei Zhang
Design of Forest Tourism Resources Evaluation System based on GIS........................171
Chuanming Yang, Xiaomei Zhang
Research on the Fluctuation of RMB Exchange Rate based on Data
Mining Method...............................................................................................................182
Ying Ye, Yirong Ying
Study on Wireless Data Transmission System based on GPRS and ARM .................. 195
Mingyu Zhang, Yuling Wang, ShuLan Gong
Research on the Appraisal Index System of Real Estate based on BP
Neural Network............................................................................................................. 207
Jia Liu, Xiaode Zuo, Mingjian Xu
Empirical Analysis on the Efficiency of Financial Governance based on
Computer-aided Data Mining ....................................................................................... 217
Chao Ma
Research on Fuzzy PID Control Strategy for Brushless Direct Current Motor ............227
Lili Jing, Hua Ge, Yang Nie
Research on the Mobile Internet Fragmentation Economic based on
Support Vector Machine ............................................................................................... 238
Hong Jiang, Hong Zhao
Research on Teaching Evaluation of College Physical Education Based on
Data Mining and Network Survey................................................................................. 249
Zhengdao Lu, Yibo Zhao
Research on the Competency Model of Innovative Entrepreneurial Team
based on Network Information Data Mining Technology............................................ 260
Jia Liu
RISTI, N.º E5, 07/2016 xi

Study on the Coordinated Control System of Urban Traffic Trunk Line
based on Particle Swarm Optimization..........................................................................273
Chengtao Cao, Feng Cui, Lunhui Xu
Extraction Process Optimization of Total Flavones from Moriga Oleifera
Leaves Based on Computer-aided Response Surface Method..................................... 287
Yi Zheng, Bin Wu, Jianmei Deng, Shaojuan Jiang
Technology Innovation and Manufacturing Upgrading Drive Mechanism
based on Internet Plus................................................................................................... 299
Feng Ji, Ting Zhang, Fan Wu, Bangjun Wang
Supply Chain Risk Assessment based on Support Vector Machine .............................310
Wei Cao, Xi Zhang
An Improved K-means Algorithm Application in Distributed Data Mining
and Information Technology......................................................................................... 323
Jianhu Gong
Research on the Formation and Development Mechanism of Brand Loyalty
based on Social Media .................................................................................................. 334
Hongxin Li
Research on Social Support System of Peasant Workers’ Citizenship
based on Data Mining Theory ......................................................................................347
Ji Liu, Xiaomei Zhang , Zhilin Suo
Research on the E-business Enterprise Employee Occupation Commitment
and Career happiness based on Network Survey ......................................................... 360
Shuliang Lv, Tiange Duan, Lanying Du
The Influence of Rural Financial Development on Urban-rural Income
Gap based on Data Mining Method...............................................................................374
Qiwen Zhang, Weiwei Lu, Jinyuan Wang
Simulation of Influencing Factors in Internet-Enabled Transformation
Performance of Traditional Enterprise Based on System Dynamics ..........................387
Feng Ji, Ting Zhang, Fan Wu, YuPeng Mou
Design of Injection Mold for Automobile Grille Based on
Computer-aided CAD Technology ..............................................................................400
Jinlian Wang
Research on Rolling Bearing Fault Diagnosis Based on Support
Vector Machine...............................................................................................................413
Shusen Li, Qingchun Zhang, Xin Shang, Bin-bin zhang
Research on Public Management Efficiency Improvement Method based on
Parallel Database Oriented Optimization Management Information System..............425
Yingjun Han, Rui Sun
xii RISTI, N.º E5, 07/2016

Adjustment and Application of Ground Motion Input in Temporal
Analysis of Structure Aseismic .....................................................................................437
Changhao Zhang, Jianhu Feng, Xuntao Wang
Design and Development of a Dynamic Wearable Computer Device Platform ...........447
Zhouping Yin
Design and Implementation of Communication System Based on
Android Platform .......................................................................................................... 459
Mingming Chen, Hong Tang, Yuzhi Chen, Ning Wang
RISTI, N.º E5, 07/2016 xiii

Revista Ibérica de Sistemas e Tecnologias de Informação Recebido/Submission: 01/03/2016
Iberian Journal of Information Systems and Technologies Aceitação/Acceptance: 04/07/2016
Research on Influence Factors of New Rural Sports

Culture Development Based on Data Mining Technology
Huasheng Huang1, Xiaojun Wu1, Ganchen Tao2, Ying Zhang2, Yuefeng Wang2,
Xiaodong Long2,*
Long Xiaodong,*171761779@qq.com
1
Guangxi University of Chinese Medicine, Nanning, Guangxi, China
2
P.E. Department, Jinggangshan University, Ji’an Jiangxi, China
Pages: 1–12
Abstract: With the advent of the information age, the Internet and people’s
production and life are increasingly inseparable. In this paper, we analyze network
culture construction in rural areas, and find out that development of network
culture has accelerated the economic development in rural areas, also brought new
vitality to the new countryside. At the same time, by using data mining method the
results show that factors of the impact of Jiangxi new countryside sports culture
development as government behavior influence factor, economic impact factor,
sports industry investment influence factor, rural influence of peasants’ sports,
environmental impact factor and social influence factor. So that, the government
should fully implement the new rural business information service system
construction, enrich the cultural life of rural sport.
Keywords: Information network, data mining, new countryside, sports culture,
E-commerce platform, rural economy
1. Introduction
With the advent of the information age, the Internet and people’s production and life are
increasingly inseparable, has penetrated all levels and in all industries. In recent years,
the Internet in China has been rapid development, according to the CNNIC released
the 27th China Internet development statistics report shows, the country rural reached
1.86 billion, accounted for 27.9% of the overall Internet users, an increase of 16.9%. At
present, China’s efforts to solve the “three rural issues”, promoting the development of
rural economy, promoting the construction of new countryside, the next period of time
will emerge more farmers in the internet(Lastra, 2015). Network culture is changing
the mode of economic growth in rural areas, rural spontaneous a typical representative
of the application of information technology completely changed the fate of peasants
and farmers through the spontaneous use of market of electronic commerce transaction
platform, direct docking market demand, drive the development of manufacturing in
rural areas and other supporting industries, promote the upgrading of rural industrial
structure and economic development (Abreu, 2015; Azadi, 2014). We are in the era of
information explosion, the rapid development of Internet makes network culture in rural
RISTI, N.º E5, 07/2016 1

Research on Influence Factors of New Rural Sports Culture Development Based on Data Mining Technology
areas into the remote mountain village, into the life of the peasants, is making efforts to
change rural for thousands of years, urban and rural information unequal status, rural
network culture has become a rural information window, has also become a bridge for the
development of rural economy. Network culture is due to the application of computers
and the development of the Internet and the rise of a new cultural style. Network uses,
content as well as the network and the human society associated with the occurrence and
the physical and nonphysical to the rule of people’s thoughts, values, norms of behavior,
and way of life reshape together constitutes the network culture (Sok,2015).
The new countryside is a concept peculiar to China, the new rural sports culture is all
related with sports culture in the sum of China’s socialist system under. The central
government issued the CPC Central Committee on deepening reform of cultural system
to promote socialist cultural development and prosperity of a number of major issues,
Jiangxi Province issued 2013-2015 cultural reform and development plan, the Jiangxi
provincial government issued the Jiangxi provincial people’s Government on speeding up
some policies and measures, cultural and creative industries at the same time pointed out:
first from the cultural infrastructure construction, the development of cultural industry,
cultural system reform and innovation, cultural talent team construction, security and
other aspects of policy measures to clear our province cultural development strategy;
secondly put forward to explore Jiangxi cultural heritage, promote the prosperity of
cultural industries ideas, put forward the protection and development of local cultural
heritage. Jiangxi is located in the south of China, elegant living environment, but the
economy is relatively backward, resulting in new rural sports culture in general is
relatively backward. How to accelerate the new rural sports culture, optimization and
development is the urgent problem, for in the countryside comprehensive construction
of a well-off society in an all-round way, optimize regional economic structure, promote
regional economic growth, strengthen the construction of spiritual civilization is of
great practical significance (Yazdanpanah, 2015; Urrea-Hernandez, 2015). At the same
time to implement the central Jiangxi Province and relevant documents, sound and
fast promote the cultural development of new rural sports, accelerated growth of sports
population in rural areas, promote the health of the peasant, improve the farmer quality,
rich farmers in the sports and cultural life has great practical significance.
2. Data mining and the rural information network

2.1. Data mining
Data mining is an important part of knowledge discovery. It is found that the data mining
is done through the establishment of the mining model and the corresponding algorithm
(Vachon, 2006; Jabbour, 2014). Nowadays, many kinds of models and algorithms of
association rules, classification, clustering, and decision tree and so on are provided in
order to select and use. Sampling is a classical statistical technique, has been studied
for over a hundred years of history, especially the random sampling technique, there
are many basic principles, and describes the effectiveness of random sampling. In the
field of data management, a small part of a subset of data sampling can capture the basic
features of the data through extraction to represent the total data set, and according to
the sample set to obtain approximate query results, or based on the data mining of the
sample set. Figure 1 is a representative sampling method of classification in data mining.
2 RISTI, N.º E5, 07/2016

RISTI - Revista Ibérica de Sistemas e Tecnologias de Informação
The probability is selected according to the data items are the same, the sampling method
can be divided into uniform sampling and sampling bias.
Figure 1 – The data mining process and influence

As the name suggests, in the uniform sampling each data item in the same probability
of being selected in the sampling bias, selected probability of different elements may be
different. Structure Lattice is often used to enumerate all possible itemsets, in general, a
data set containing d different items may produce R rule, R can be expressed as follows:
d −1 
 d  d −k  d − k 
R =∑   × ∑  d d +1
  =3 − 2 + 1 (1)
= k 1= k
  j 1  j 
Apriori algorithm is mining association rules is an important method, by using the
candidate item sets to find frequent item set. If the set {x, y} satisfy “support” conditions,
then each element in the collection should also occur frequently. The composition of the
set can be represented by a tree, as shown in Figure 2.
Figure 2 – Itemsets transcendental principle
RISTI, N.º E5, 07/2016 3

Assume itemsets is non-frequent, this kind of pruning index search space based on the
support measure is called the support degree based pruning technology, as shown in
Figure 3
Figure 3 – Support-Based Pruning
2.2. The construction of rural information network

The foundation of rural informatization construction is the construction of information
infrastructure. The network infrastructure of rural informatization is based on the basic
conditions and guarantees the implementation of rural informatization, rural information
network is an important tool to realize rural informatization. Rural information network
coverage is the most basic problem that must be solved in the development of village
information. Governments at all levels should be on the existing basis, to further increase
investment, and strengthen the construction of rural infrastructure, vigorously carry out
and promote the application of various network technologies (Michelsen,2006). The
construction of the rural information infrastructure because of small scale, decentralized
investment, the central and provincial governments to direct operation, need to rely on
local government and grass-roots organizations to implement(Asudi, 2015).
1. Network culture can effectively promote rural labor productivity: the development
of the Internet makes the world smaller; the Internet will become an important
engine for leading the development of rural economy. Especially the advanced
productivity and business model represented by electronic commerce, provides a
new direction for solving the “three rural issues”. Farmers through the Internet,
learning to advanced farming technology, can learn a variety of science and
technology to become rich valuable experience, can accurately and timely know
the city unit recruitment information, farmers can directly on the net release of
agricultural products sales information, can effectively reduce the circulation link,
reduce costs, increase revenue (Vande ,2015).
4 RISTI, N.º E5, 07/2016

2. Rural network culture constructions are an important platform to improve the

quality of farmers: at present, the degree of rural areas is still very low. network
cultural construction in rural areas can not only provide intellectual support and
spiritual impetus for the development of rural economy, the more important is
to be able to provide a successor force of an endless supply of the in-depth reform
of the rural. Only to strengthen the network of cultural construction in rural
areas to establish a new socialist culture, so that the overall level of civilization
construction in rural areas have a larger increase, to further accelerate China’s
rural modernization.
3. The development of network culture to change the psychological status of farmers
have great effect: from peasant” variable, especially after 80 some no longer
rely on farming for a living, a new generation of farmers in their psychological
change farmer vulnerable shadow. These new generation of farmers learned to
surf the Internet directly from the Internet learn a lot of useful knowledge, in
living habits is already standard of urban people, from their minds the “farmers”
burden completely get rid of, this is probably network culture construction in
rural area of rural economic development the most practical significance:
network information into the countryside, the farmers do not envy city people,
can feel at ease to building a new socialist countryside.
3. The research object and methods

3.1. Research objects
Expert questionnaire survey of national sports culture of 16 experts and leaders;
Jiangxi province six city (Ji’an City, Ganzhou City, Nanchang, Jiujiang and Pingxiang
City and Fuzhou City) as 223 experts, the relevant leaders and rural farmers
were questionnaire.
3.2. Main research methods

•• Literature: topic as much as possible full access to domestic and foreign
research literature, the absorption and digestion of the achievements of the
relevant literature, understand the theory of the frontier and progress, especially
by foreign scholars use the theory and method, drawing on existing research
results and research methods, the theoretical framework and hypotheses are
proposed and measurement for the empirical study in this paper the concept of
looking for a useful reference.
•• Comparative analysis: comparison of human understanding, and to
determine the relation between the differences between things, and reveal the
most common way of thinking of the nature of things.
•• Expert interviews: interviews is the main conception of this study are
validated and the further adjustment, and as a measure of the conceptual model
and research the variable table and draw lessons from the experience of experts
and relevant personnel, for the index of validity of the guarantee to determine
the specific content.
•• Questionnaire: questionnaire survey is an important method for the research
of this study, through the discussion on the structure of the empirical study on
RISTI, N.º E5, 07/2016 5

variable indicators measuring scale and in formal empirical research by larger

scale questionnaire validation study ideas and research hypothesis.
•• Mathematical statistics: mainly used to determine the value of the collected
data, verify the hypothesis. According to the aim of the study, this study using
several statistical analysis methods of the questionnaire survey to collect data
processing, examine the various concepts of measurement scale and verify
hypotheses proposed in the research model. This research mainly adopts two
kinds of statistical analysis tools of SPSS16.0 software and EXCELL statistical
package for the social sciences.
•• Logical reasoning: by using comparison, deduction, induction, analysis,
reasoning, logical analysis of the results, draw conclusions from.
4. Results
4.1. The impact factors of rural sports culture
Article employ 16 experts or leadership questionnaire, 32 Factors Affecting factors issues
to statistical factor analysis, Jiangxi cultural factors affecting rural sports factor collation,
analysis and classification. Based on the KMO value and Bach Wright shows: KMO =
0.832; Bach Wright test value X2=146.639, P=0.000<0.01; there was a significant
difference in the factor analysis model of data for statistics. Table 1 results show that
remove “Kaiser” eigenvalues greater than 2 common factors, a total of six common main
factor, followed by the contribution rate as 18.387%, 15.480%, 12.859%, 11.411%, 9.488%,
7.457%; cumulative contribution rate =75.082%>70%. Due to the construction of rural
sports culture elements is extremely complex, there are a large number of elements
without development and dig out, according to the statistical significance, can reflect the
overall amount of information.
Initial Eigenvalues Rotation Sums of Squared

Component Loadings
Total Variance Cumulative % Total Variance Cumulative %
F1 5.884 18.387 18.387 4.595 14.361 14.361
F2 4.953 15.480 33.867 4.419 13.808 28.169
F3 4.115 12.859 46.726 4.263 13.323 41.492
F4 3.652 11.411 58.137 3.870 12.095 53.587
F5 3.036 9.488 67.625 3.538 11.058 64.645
F6 2.386 7.457 75.082 3.340 10.437 75.082
Table 1 – Total Variance Explained
4.2. Classification of factors

Construction of the main factors influencing factors of rural sports culture is
very complex, we use statistical factor analysis of variance orthogonal rotation to
maximize its impact factor to give six main factors reasonably classify and name (as
shown in Table 2).
6 RISTI, N.º E5, 07/2016

Component
indicators
F1 F2 F3 F4 F5 F6
1 government organizations influence 0.584
factors
2 rural sport participation factors 0.528
3 rural sports popularity factors 0.816
4 rural sports participation factors 0.555
5 central government investment 0.801
6 Rural Sports Association 0.642
7 rural sports activities of the organization 0.938
8 influence diversity of rural sports 0.617
9 rural sports entertainment function 0.658
10 factors affecting rural climatic 0.557
conditions
11 factors influence the rural area 0.562
12 central government’s fiscal 0.763
13 factors influence rural sports equipment 0.564
14 innovation of rural sports project 0.686
15 local government investment factors 0.686
16 local government fiscal economic 0.641
investment
17 rural sports ornamental function factors 0.825
18 local rural economic development level 0.642
19 local rural sports characteristics 0.491
20 development of rural sports history 0.837
21 Sports and cultural factors 0.892
22 factors of government management 0.727
23 factors of government management 0.629
24 local rural economic development 0.793
25 rural sports organization personnel 0.621
26 degree of government support 0.858
27 farmers’ per capita income level 0.799
28 local social and political environment 0.777
29 rural sports culture dissemination 0.554
30 rural sports industry 0.600
31 individual investment of rural sports 0.717
32 government propaganda effect 0.554
Table 2 – Component Matrix
RISTI, N.º E5, 07/2016 7

So that, 32 factors influencing factors can be summarized as: “The government Behavior
Factors Factor”, “economic factors factor”, “sports industry input factors Factor”, “sports
and cultural life of rural farmers factors factor”, “rural area outside environmental
factors “and” social factors influence factor “and other six types.
4.3. New development strategy of sports culture

By 223 experts, relevant leaders and farmers in rural areas, to investigate the government
behavior index important degree view, input SPSS16.0 statistical software for processing.
The results in Table 3 show: the importance of government action, government organization
metric, the mean and standard deviation (3.6637 ± 0.56861), T = 96.218 and P = 0.000;
the importance of Physical Education Association, indicators of government action, mean
and standard deviation (4.6547 ± 0.50409), T = 137.890 and P = 0.000; in government
action “government official” indicators in importance, mean and standard deviation
(3.6278 ± 0.65796), T = 82.337 and P = 0.000; the importance of government action
rural Social instructor indicators, the mean and standard deviation (4.1749 ± 0.74186),
T = 84.038 and P = 0.000; in government action sports organization key personnel, the
importance of indicators, the mean and standard difference (3.4843 ± 0.64965), T =
80.092 and P = 0.000; degree of importance in government action “level of government
support” index, the mean and standard deviation (4.5291 ± 0.57563), T = 117.496 and
P = 0.000; in government action the importance of “social and political environment”
indicators, the mean and standard deviation (3.9910 ± 0.75927), T = 78.495 and P =
0.000; the importance of government action, government propaganda indicators, the
mean and standard deviation (2.8969 ± 0.67992), T = 63.624 and P = 0.000. Government
to promote the importance of behavioral indicators Jiangxi Rural Development of Sports
Culture optimization were: sports associations; the degree of government support; rural
social instructors; government propaganda; sports organizations key personnel; social
and political environment; governmental organizations; government managers.
indicators Mean value T value P value ranking

Government organization 3.6637±0.56861 96.218 0.000 5
Sports Association 4.6547±0.50409 137.890 0.000 1
government management staff 3.6278±0.65796 82.337 0.000 6
rural social instructor 4.1749±0.74186 84.038 0.000 3
Sports organization key staff 3.4843±0.64965 80.092 0.000 7
Government support 4.5291±0.57563 117.496 0.000 2
social and political environment 3.9910±0.75927 78.495 0.000 4
government propaganda 2.8969±0.67992 63.624 0.000 8
Table 3 – The Jiangxi government behavior index(n=223)

From Table 4, the results show that: in all areas of the economy, the central government
investment in the economy, the importance of indicators, the mean and standard deviation
(2.9238 ± 0.71567), T = 61.008 and P = 0.000; In the “Local Government Finance and
Economy input” indicators the degree of importance, the mean and standard deviation
(3.4664 ± 0.65587), T = 78.924 and P = 0.000; degree of importance “level of local rural
8 RISTI, N.º E5, 07/2016

economic development” indicators, the mean and standard deviation (3.4215 ± 0.53850),
T = 94.882 and P = 0.000; degree of importance, “per capita income level of farmers’
indicators, the mean and standard deviation (3.6502 ± 0.64626), T = 84.347 and P = 0.000;
in farmer sports personal investment an important indicator of the degree of mean standard
deviation (4.1883 ± 0.69778) and, T = 89.635 and P = 0.000; in the central government
investment in sports infrastructure indicators of the degree of importance, the mean and
standard deviation (4.6054 ± 0.48987), T = 140.391 and P = 0.000 ; the importance of
local government funding for sports activities in rural areas metric, the mean and standard
deviation (4.0045 ± 1.05479), T = 56.693 and P = 0.000; in various economic fields “physical
and cultural quality of rural farmers’ index the degree of importance, the mean and standard
deviation (3.6054 ± 0.51672), T = 104.196 and P = 0.000; degree of importance, the level
of development of rural sports industry indicators, the mean and standard deviation
(4.5874 ± 0.49340), T = 138.843 and P = 0.000. Indicators of the rural economy in various
fields to promote the importance of optimizing the development of Jiangxi Rural Sports
Culture in the order, the central government sports infrastructure investment; the level of
development of rural sports industry; farmer sports personal investment; the financing of
local government investment in rural sports activities; farmers annual per capita income
levels; physical and cultural quality of rural farmers; local government finances economic
investment; the level of the local rural economic development; the central government
financial and economic investment.

central government’s fiscal 2.9238±0.71567 61.008 0.000 9
local government financial investment 3.4664±0.65587 78.924 0.000 7
local rural economic 3.4215±0.53850 94.882 0.000 8
annual per capita income of farmers 3.6502±0.64626 84.347 0.000 5
farmer sports personal investment 4.1883±0.69778 89.635 0.000 3
central government investment 4.6054±0.48987 140.391 0.000 1
local government investment 4.0045±1.05479 56.693 0.000 4
Rural sports culture quality 3.6054±0.51672 104.196 0.000 6
development of rural sports industry 4.5874±0.49340 138.843 0.000 2
Table 4 – Jiangxi rural economic index important degree (n=223)

From Table 5, the results show that: the importance of sports and cultural life of Jiangxi Rural
farmers “Farmers sports participation” index, the mean and standard deviation (4.0807 ±
0.63865), T = 95.418 and P = 0.000; in “farmer sports liking” metric the degree of importance,
the mean and standard deviation (3.8610 ± 0.79608), T = 72.426 and P = 0.000; degree
of importance, “government official” indicators, the mean and standard deviation (3.5291
± 0.52659), T = 100.080 and P = 0.000; degree of importance “farmers sports diversity”
indicator, the mean and standard deviation (4.0628 ± 0.82506), T = 73.534 and P = 0.000;
the importance of “farmer sports watch features” index, the mean and standard deviation
(3.7848 ± 1.04767), T = 53.947 and P = 0.000. Farmers living index Sports Culture to promote
the importance of optimizing the development of Jiangxi Rural Sports Culture in the order
of: the extent of the farmers participating in sports; sports diversity of farmers; farmer sports
liking; farmer sports watch function; government managers.
RISTI, N.º E5, 07/2016 9


Farmers’ participation 4.0807±0.63865 95.418 0.000 1
Farmers love sports 3.8610±0.79608 72.426 0.000 3
Government management personnel 3.5291±0.52659 100.080 0.000 5
Diversity of peasant sports project 4.0628±0.82506 73.534 0.000 2
Sports Watch features 3.7848±1.04767 53.947 0.000 4
Table 5 – Jiangxi rural sports culture life index (n=223)

The results in table 6 shows the importance of Jiangxi Rural external environment “Rural
climatic conditions” indicators, the mean and standard deviation (3.4484 ± 0.52486),
T = 98.113 and P = 0.000; in “rural site conditions” indicators of the degree of importance ,
mean and standard deviation (2.8251 ± 0.74186), T = 56.868 and P = 0.000; degree
of importance “rural sports equipment” indicators, the mean and standard deviation
(4.5964 ± 0.49172), T = 139.590 and P = 0.000 ; the importance of “history of rural
Sports development” indicators, the mean and standard deviation (4.3363 ± 0.70320),
T = 92.087 and P = 0.000; the importance of “other level cultural development in rural
areas” metric, the mean and standard deviation (3.6771 ± 0.88214), T = 62.248 and
P = 0.000. External environmental indicators to promote the importance of rural
Jiangxi Rural Sports Culture optimize the development were: Rural sports equipment;
sports history of rural development; other levels of cultural development in rural areas;
rural climatic conditions; rural site conditions.

Rural climatic conditions 3.4484±0.52486 98.113 0.000 4
The rural site 2.8251±0.74186 56.868 0.000 5
Rural sports equipment 4.5964±0.49172 139.590 0.000 1
development of rural sports history 4.3363±0.70320 92.087 0.000 2
rural cultural development 3.6771±0.88214 62.248 0.000 3
Table 6 – Jiangxi rural environment index (n=223)

Table 7 show that the importance of social influence in rural Jiangxi Province, “the
number of rural sports” indicators, the mean and standard deviation (4.0448 ±
0.60609), T = 99.659 and P = 0.000; in “rural sports organization” metric the degree of
importance, the mean and standard deviation (2.7534 ± 0.54291), T = 75.733 and P =
0.000; in “rural sports innovation” the importance of indicators, the mean and standard
deviation (3.6188 ± 0.64585), T = 83.674 and P = 0.000; the importance of “local rural
sports characteristics” indicators, the mean and standard deviation (3.5516 ± 0.55814),
T = 95.024 and P = 0.000; degree of importance “rural Sports culture Communication”
metric, the mean standard deviation (4.5516 ± 0.51621) and, T = 131.670 and P =
0.000. Rural indicators of social influence to promote the importance of optimizing
the development of Jiangxi Rural Sports Culture were: rural sports culture spread; the
number of rural sports; sports innovation in rural areas; local rural sports activities
characteristics; the number of organized sports activities in rural areas.
10 RISTI, N.º E5, 07/2016


number of rural sports 4.0448±0.60609 99.659 0.000 2
number of rural sports organizations 2.7534±0.54291 75.733 0.000 5
Rural sports innovation 3.6188±0.64585 83.674 0.000 3
characteristics of rural sports 3.5516±0.55814 95.024 0.000 4
spread of sports culture 4.5516±0.51621 131.670 0.000 1
Table 7 – Jiangxi rural social influence index(n=223)
5. Conclusion
Network information industry has become a new bright spot in China’s economic
growth, and penetrates into the whole process of agricultural modernization,
influence on the development of agricultural economy is more and more deep, play
an important role. The report pointed out that accelerating the transformation of
economic development, the development of the information industry.. However, we
should also see that although the pace of development of China’s Internet market is
very rapid, but security is conducive to rural network culture market development
policy and system lags behind and support. Therefore, the government should increase
the pace of information construction in the new rural construction, in the rural areas
to set up a network of specialized services, township institutions can set up a network
information service center, directly for the “three rural”, set up public cafes in rural
residential area. This model helps to absorb the network management personnel,
can effectively promote the employment of university students, also the “three rural
service”. Empirical results show that the effect of new rural sports culture of Jiangxi
optimization factors for the development of a: “the government behavior influence
factor”, “economic impact factor”, “sports industry investment influence factor”, “rural
farmers sports culture life factor”, outside of the rural areas of environmental impact
factor “and” social influence factor “six types. Optimization of sports culture in rural
areas of Jiangxi new development strategies are: change the behavior of the government,
increase in all areas of economic input, rich in Jiangxi rural sports and cultural life of
farmers, beautify external environment in rural areas of Jiangxi, harmony and stability
of Jiangxi rural social influence.
Acknowledgments
This study was financially supported by Jiangxi Academy of social science and art project
planning(YG2015148).
References
Abreu, A., Rocha, Á., Cota, M. P., & Carvalho, J. V. (2015). Caderneta Eletrónica
no Processo Ensino-Aprendizagem: Visão de Professores e Pais de alunos do
ensino Básico e Secundário. RISTI-Revista Ibérica de Sistemas e Tecnologias de
Informação, (16), 108–128.
RISTI, N.º E5, 07/2016 11

Asudi, G. (2015). Napier grass stunt disease in East Africa: Farmers’ perspectives on
disease management. Crop Protection, (71), 116–124.
Azadi, M., Shabani, A. (2014). Planning in feasible region by two-stage target-setting
DEA methods: An application in green supply chain management of public
transportation service providers. Transportation Research Part E: Logistics and
Transportation Review, (70), 324–338.
Jabbour, A. (2014). Mixed methodology to analyze the relationship between maturity of
environmental management and the adoption of green supply chain management
in Brazil. Resources. Conservation and Recycling, (92), 255–267.
Lastra, B. X., Hubbard, C. (2015). What drives farmers’ participation in EU
agri-environmental schemes? Results from a qualitative meta-analysis.
Environmental Science & Policy, 54, 1–9.
Michelsen, O., Magerholm, A. (2006). Eco-efficiency in extended supply chains:
A case study of furniture production. Journal of Environmental Management, (79),
290–298.
Sok, J., Hogeveen, H., Elbers, A.R. (2015). Farmers’ beliefs and voluntary vaccination
schemes: Bluetongue in Dutch dairy cattle. Food Policy, (57), 40–49.
Urrea-Hernandez, C., Almekinders, C. (2015). Understanding perceptions of potato
seed quality among small-scale farmers in Peruvian highlands. NJAS - Wageningen
Journal of Life Sciences, (40), 95–98.
Vachon, S. (2006). Green project partnership in the supply chain: the case of the package
printing industry. Journal of Cleaner Production, (14), 661–671.
Vande F., Claerebout, E. (2015). Diagnosis before treatment: Identifying dairy farmers’
determinants for the adoption of sustainable practices in gastrointestinal nematode
control. Veterinary Parasitology, (212), 308–317.
Yazdanpanah, M., Feyzabad, F. (2015). Predicting farmers’ water conservation goals and
behavior in Iran: A test of social cognitive theory. Land Use Policy, (47), 401–407.
12 RISTI, N.º E5, 07/2016

Influence of Corporate Equity Incentive and CSR on

Firm Performance: an Empirical Analysis based on
Large-scale Data Mining
Ming Li 1, Zhenxiang Chen 1,2,*
*zhenxiangchen123@gmail.com
1
School of Management, Wuhan Technology and Business University, Wuhan 430065, China
2
Hubei Business Service D&R Center, Wuhan 430065, China
Pages: 13–25
Abstract: According to the stakeholder theory, employees are important

stakeholders of the enterprise. Enterprises ignore the social responsibility of
employees may bring to the enterprise staff turnover rate increased, the risk of
recruitment difficulties. Therefore, there are a large number of enterprises to start
the implementation of employee stock ownership incentive system. In this paper,
by using data mining method, we choose a stock incentive in China’s 2006-2012
shares listed companies, then use paired samples T test and cross section regression
analysis of equity incentive effects and external variables on the effect of the effect.
The research shows that the corporate performance has a significant growth
before and after the stock ownership incentive, and the employee stock ownership
incentive has a positive effect on the enterprise performance, but this effect is more
significant in private enterprises and small scale enterprises. Equity concentration
is the positive effect of the employee stock ownership incentive effect, while the
macro economy has a positive effect on the positive effect of equity incentive.
Keywords: Data mining, employee shareholding, equity incentive, mechanism
analysis, corporate social responsibility
1. Introduction
According to the stakeholder theory, employees are important stakeholders of the
enterprise; corporate social responsibility to employees may be the risk of rising
employee turnover rate, recruitment difficulties. Therefore, there are a large number of
enterprises to start the implementation of employee stock ownership incentive system.
Employee stock ownership incentive is a modern enterprise of a strategic management
decision-making behavior, when the business owners and the separation of ownership
and management (i.e. the principal agent relationship), in order to ensure that operators
can maximize play professional level, and ensuring that it does not damage the interests
of the enterprise, allowing employees to hold shares or options of the enterprise in order
to solve the information asymmetry problem in corporate governance. The development
of capital market in western countries as, equity incentive has been widely used, such
as one of the earliest American Equity Incentives mainly used in Silicon Valley and
RISTI, N.º E5, 07/2016 13

Influence of Corporate Equity Incentive and CSR on Firm Performance: an Empirical Analysis based on Large-scale Data Mining
other knowledge innovation enterprises and high-tech enterprises, experience shows

that the long-term incentive of enterprise managers and employees showed better
effect. Since then Japan, Britain, France, Singapore began to follow, equity incentive
system obtained the good development, at present in the global top 500 enterprises,
85% to take equity incentive system, said this system not only an effective solution to the
corporate governance in modern capital market, optimizing the relationship between
the shareholder (owner), managers and outside investors.
Our country in the capital market has just set up was an early study of equity incentive
in listed companies, but ultimately failed to perform, until 2005, equity division reform,
the relevant departments of the state introduced equity incentive management approach
China Securities Regulatory Commission in December 2005 promulgated the “listed
company equity incentive management approach”, the Ministry of Finance in February
2006 released the state-owned holding listed companies the implementation of equity
incentive pilot scheme, the listed companies in the implementation of equity incentive
to clear away the obstacles, from China’s employee stock ownership incentive entered
the phase of rapid development. But with the increase of professional managers, and
enlarge the scale of enterprise, employee stock option motivation to the development
of the enterprise, whether there is the significant effect began to be questioned, many
scholars think due to equity incentive targets are senior corporate management
personnel, these personnel not only stock right incentive, and get higher pay, and a few
years China’s economic development better, and therefore ability general managers can
also basically realize motivating goal set by the owner of the, not reflect the level of
incentive objects. In addition, the agency governance will also lead to moral problems,
managers to achieve incentive targets, the preparation of false financial statements or
spread false information, damage the long-term growth of the listing Corporation and
the interests of small and medium investors.
In the existing theoretical research, the study of employee stock ownership incentive is
mainly focused on whether there is the incentive effect, the direction of the incentive
effect, the strength of the. In the incentive effect, it is divided into direct effect and indirect
effect. Direct effects including the effects of the three directions, as the effect based
on the convergence of interest hypothesis that agent (operators) equity incentive will
expand its consistency and interests among the shareholders, can stimulate the behavior
of performance management, finally bring forward the enterprise performance(Abreu,
2015). Entrenchment effect hypothesis pointed out that operators of rent-seeking
behavior, equity based incentives may be because of the abuse of executive and cannot
effectively alleviate the agency conflicts, but will exacerbate agency problems, is not
conducive to enterprise performance. Also a point of view that is a nonlinear relationship
between equity incentive and corporate performance, when the incentive object number
of shares less than a certain threshold, the positive incentive effect, and the number of
shares more than a certain threshold, negative incentive effects (Shen, 2011; Azadi, 2014).
For the indirect effect, it is mainly of equity incentive by what means to improve enterprise
performance, managers will reduce dividend payments by expanding investment and
increase earnings management, raise the innovation strength to promote enterprise
performance improvement, so as to realize the incentive effect. Currently more scholars
from the sample data of listed companies of, the equity incentive effect empirical research,
14 RISTI, N.º E5, 07/2016

such as Huang (2009) resaerch listed companies section data found that employee stock
ownership on firm performance has a positive effect, but the management shareholdings
influence on firm performance is 4-6 times the ordinary employee stock ownership, but
impact on enterprise performance and employee stock ownership there is an inflection
point. Kong (2015) using in Shaanxi Province in the unlisted small and medium sized
enterprise data analysis found that employee stock ownership plans to significantly
improve the corporate performance, the incentive effect of the employee stock ownership
plan exist boundary conditions, employees holding number more, the incentive effect
of the employee stock ownership plan smaller and smaller. Wang (2015) during the
period 2006-2010 implementation of equity incentive of listed companies as a sample,
the study found and non-implementation of the equity incentive plan of companies
compared to the implementation of equity incentive plan of company performance
before and after the implementation in are good, and in the non state owned enterprises
is more significant. There are also scholars have studied the effects of external variables
of the incentive effects, Chen (2015) study the effect of state ownership and market
competition on the relation between equity incentive and corporate performance, found
that the equity incentive have incentive effect, but the state-owned holding weakened
the effect of the implementation of equity incentive, product market competition will
strengthen the positive relationship between equity incentive and firm performance.
Wu (2014) analysis of the influence of board characteristics on the governance effect
of equity incentive, the study found that the equity incentive policy produced a positive
incentive effect, the board of directors shareholding and the proportion of independent
directors increase can enhance incentive effect, and the chairman and general manager
together while reduce the incentive effect.
This paper chooses Chinese manufacturing listed companies as the analysis object,
compared to the previous literature, there are two aspects of innovation, using
cross-sectional data, which dependent variable is the equity incentive and corporate
performance change amount and on the amount of changes whether there significant
difference of paired sample t test analysis; second influences the effect of equity
incentive factors, mainly including two aspects of the ownership structure of the firm
and macroeconomic environment, the former refers to the internal characteristics of the
company, the latter points to the external environment.
2. Large-scale data mining

2.1. BIRCH algorithm
In general, the single pass scan data set can get better clustering results. Optional follow-
up of multiple scan can further improve the quality of clustering. From the statistical
point of view, clustering is given a consists of N data points in the data set and a distance
measuring functions f (such as computing clusters within each of the two data points
between the average distance function), the data set is divided into K clusters (or
not given number k, by the algorithm automatically discover the optimal number of
clusters), final result is to find a for a data set of the optimal partition, which makes the
distance measure function f of the minimum value. From the point of view of machine
learning, clustering is an unsupervised learning algorithm, by clustering data into n
RISTI, N.º E5, 07/2016 15

clusters, which minimizes the distance between points within the cluster. The core of
BIRCH method in the micro clustering stage is the construction of CF tree as follows:
CF = ( n, LS , SS ) (1)
n
LS = ∑ xi (2)
i =1
n
SS = ∑ x 2i (3)
i =1
The use of CF (cluster feature) can be derived from a part of Statistics: c

n
∑x i
LS (4)
x0 = i =1
=
n n
∑( x − x0 )
2
i (5)
nSS − 2 LS 2 + nLS
R= i =1
=
n n2
n n
∑ ∑( x − xj )
2
i =1 j =1
i
2nSS − 2 LS 2 (6)
d= =
n (n − 1) n (n − 1)
The average distance of the objects in the cluster was measured by d.
SS1 SS2 2 LS1 LS2

d* = + − (7)
N1 N2 N1 N2
2.2. Large-scale data mining

In recent years, data mining has attracted great attention of the information industry,
the main reason is the existence of a large number of data, can be widely used, and the
urgent need to convert these data into useful information and knowledge. Access to
information and knowledge can be widely used in a variety of applications, including
business management, production control, market analysis, engineering design and
scientific exploration. Data mining generally refers to the process of searching for
information hiding in the data from a large number of data. Data mining usually
and computer science, and through statistics, online analysis processing, information
retrieval, machine learning, expert system (rely on past experience rule) and
pattern recognition, and many other methods to achieve these goals. Data mining
classification as shown in figure 1.
16 RISTI, N.º E5, 07/2016

Figure 1 – Data mining classification

The purpose of data mining is to find more high-quality users from the data. Then I
continue to explore the guiding methods of data mining model. What is the guidance of
the data mining method model, and how to build a model of data mining is important. In
constructing a supervised data mining model, first of all to understand and define some
models that attempt to estimate the target variables, data mining process as shown in
figure 2.
Figure 2 – Data mining model
3. Model design and variables

3.1. Model building
Currently research on employee stock ownership incentive effects of equity includes the
event study and regression model method, which event study method is test incentive
behavior impact on stock prices in the short term, such as by using the event study found
RISTI, N.º E5, 07/2016 17

that listed company equity incentive announcement before and after the issuance of the
stock has a significantly positive price effect. The regression model method is mainly to
study the influence of the institutional behavior on corporate performance under the
background of employee stock ownership and stock ownership incentive. In this paper,
we mainly observe the direction and extent of the financial effect of employee stock
ownership. Effect of the equity incentive effect more is the use of panel data analysis,
namely the return of employees or executives Shareholding on firm performance
(indicators). However, the paper considers that hasn’t been properly chosen such that the
equity incentive, executive stock ownership may not be used to motivate, the body that
contains equity. In this paper, we choose the cross section data analysis, each enterprise
to select the year of the incentive value, as a result of changes in the performance of
enterprises before and after the incentive. Specific models are constructed as follows:
(1)
Among them is the enterprise performance index, is the equity incentive agent variable,
is the control variable, is the constant.
For model (1), if α is significantly positive, that equity incentive has a significant positive
impact on enterprise performance, that is, the existence of positive equity incentive
effect, if α is negative, it shows that the equity incentive has a significant negative impact
on corporate performance, that is, the existence of negative equity incentive effect, if
the coefficient α is not significant, it shows that there is no incentive effect. To further
explore the third variables of corporate equity incentive effect of, in equation (1) joined
the corporate ownership structure with equity incentive proxy variable of the cross term,
macroeconomic environment and equity incentive proxy variable of the cross terms, the
model are as follows:
(2)
(3)
Among them is the enterprise equity index, is the macroeconomic indicators, and the
rest of the meaning of the variables with the formula (1). is the cross terms of
enterprise equity incentive and enterprise equity concentration, and it is the cross terms
of enterprise equity incentive and macro economy.In the equation (1), the purpose of
introducing the cross term is to examine the influence of internal variables (ownership
structure) and external environment (macro economy) on the incentive effects of
employee stock ownership,This is mainly to observe the coefficient β. In this paper, the
least square method (OLS) is used to estimate the above equations.
3.2. Sample selection and variable

In this paper, we select A shares of Listed Companies in China during the period of
2006-2012, the reason why not include samples during the period of 2013-2014,
18 RISTI, N.º E5, 07/2016

The main consideration is that after 2013, the incentive cannot reflect the changes in
the performance of the company’s equity incentive. In order to ensure the validity of the
data, this paper removed the *ST name of the company in the year,
Excluding the three years before and after the equity incentive, this paper mainly takes
into account the changes in the performance of the stage (three years) before and after
the equity incentive. Finally get 227 enterprises cross section data. Enterprise data are
derived from the Tai’an Shenzhen financial database (CSMAR), economic sequence data
from the 2015 “China Statistical Yearbook”.
1. Enterprise performance: Enterprise performance is refers to the company
business results in a certain period of time, the modern enterprise system to
measure the enterprise’s performance there are many variables, including total
return on assets rate (ROA), net assets income rate (ROE), operating margins,
earnings per share (EPS), profit growth rate etc.. In the incentive effects of
equity research in the literature, many performance appraisal chose indicator
associated with net profit, Dong Lijuan (2013) of 150 with equity incentive of
enterprise statistics showed that, 82% of the assessment indexes of enterprises
net profit growth rate, 72% chose the weighted average return on net assets rate,
in addition to 21.3% of the enterprises chose net profit, 12.67% of the enterprises
chose operating income growth rate. After drawing on the existing literature, this
paper choose the growth rate of net profit as the index of enterprise performance
evaluation, which is the proxy variable of the majority of the study of employee
stock ownership incentive literature.
2. Corporate equity incentive(esop): the equity incentive, many literatures
have choices of executives shareholding accounted for accounted for the total
capital ratio of employee stock ownership or total equity ratio measure, this
paper argues that this part of the employee stock ownership is not real incentive
properties of, part of the shares in the former excitation have held. Therefore, this
article choose listed companies to equity incentive for the first time after notice
(sent) stake as an employee stock ownership plan index of equity incentive.
3. Ownership structure of enterprises: Ownership structure in the narrow
sense is the equity structure, but also refers to the degree of concentration
of shares in enterprises. It is generally believed that the more concentrated
ownership, is the largest shareholder of enterprise management control and
decision-making more assertive, which will increase the risks and benefits of
enterprises. In this paper, we choose the proportion of the top five shareholders
and measure the ownership structure.
4. Macro economy: This paper chooses the growth rate of gross domestic
product (GDP) to measure. For the variable for the effect of equity incentive effect
to explain for the macro economy and better that may lead to equity incentive
effect is getting better and better, but the “upside” effect exists “distortion”.
5. Control variable: This paper selects two variables as the control variables of
the enterprise scale and the assets and liabilities structure. Enterprise scale use
numerical representation of business income. Assets and liabilities structure
use the asset liability ratio (total liabilities / total assets), the greater the asset
liability ratio, indicating that the stronger the business use of financial leverage.
RISTI, N.º E5, 07/2016 19

4. Empirical analysis
4.1. The comparison of enterprise performance before and after the stock
ownership incentive
Before using the regression model analysis, firstly, the sample t test of sample
enterprises employee ownership equity incentive and net profit growth rate differences
were analyzed. The former excitation is refers to enterprises in the released incentive
announcement, i.e., motivation, the implementation of the annual) in front of three
years, excitation is refers to the enterprise in released incentive announcement back
three years (including notice that year). The specific paired samples T test results are
shown in table 1. From table 1 can see, 227 enterprises in equity incentive three years
ago, the average net profit growth rate is 16.55%, in the three years after the equity
incentive, the average net profit growth rate is not 24.10%, the second difference 7.55%.
T test statistic value is 2.033, corresponding to a bilateral probability sig values for
0.043<0.05 that China’s A-share listed companies in equity incentive and net profit
growth rate has increased significantly.
net profit
sample difference T DF Sig. (bilateral)
growth rate
Before excitation 0.1655
0.0755 2.033 226 0.043
After excitation 0.2410
Table 1 – Paired sample T test before and after equity incentive
4.2. Test on the effect of stock incentive of Listed Companies in China

Table 2 shows the cross-sectional data and regression estimation results, including
(1) is the estimation results for the full sample enterprises, column (2) - (4) is the sub
sample estimation results, including private and state-owned enterprises Division
of the reference in Taian database of actual control of corporate division standard,
enterprise scale division is on business income points for large-scale enterprises
(incentive year revenue 10 billion yuan or more) and small scale enterprises
(incentive year revenue 10 billion yuan). Table 2 columns (1) for the full sample
enterprise estimation results show that goodness degree was 0.667, observation of
all the independent variables, the employee stock ownership plan equity incentive
variable coefficient for 1.337>0 in 0.01 probability level statistically significant, that
equity incentive has a positive promoting role on the performance of enterprises
growth, this result and table 1, consistent with the conclusion. For control variables,
you can see the enterprises operating income variable coefficient for -1.862<0, at
1% significant level of probability of statistically significant, indicating that larger
enterprises, enterprise performance growth is low and even negative growth; assets
and liabilities rate variable coefficient for 10.293>0 in 5% significant probability
level statistically significant, indicating that the greater the proportion of corporate
debt and corporate performance growth is larger.
To observe sub sample estimation results, we can see private enterprises sample ESOP
variable coefficient for 7.662>0, at 1% significant level of probability of statistically
20 RISTI, N.º E5, 07/2016

significant, so for private enterprises, the implementation of employee stock ownership

equity incentive can and positive impact on firm performance, and for the state-owned
enterprises sample, variable coefficient of ESOP 0.437, in significant 10% probability level
but not significant, indicating that state-owned enterprise employee stock ownership
incentive effect is not significant. For the control variables, the results are not consistent
with the full sample, but the coefficients are the same.
For different size of enterprise, small sample sizes of ESOP variable coefficient for
1.729>0, at 1% significant level of probability of statistically significant, so for the small
scale of the enterprise, employee stock ownership plan for enterprise performance have
positive effect of equity incentive, for large-scale enterprises, the variable coefficient
of ESOP was 0.560, is not significant at the 10% level of probability that large-scale
enterprises employee stock ownership incentive effect is not significant. For the control
variables, the coefficient of symbols and was basically consistent with the previous
column, here no longer repeat one.
dependent variable performance

sample sample privately operated state-owned small-scale large-scale
column (1) (2) (3) (4) (5)
1.337*** 7.662*** 0.437 1.729*** 0.560
esop
(0.000) (0.000) (0.177) (0.001) (0.162)
-1.86*** -5.665** -0.887 -2.541** -1.634
lnincome
(0.010) (0.014) (0.227) (0.025) (0.362)
10.293** 26.600 8.660** 12.551* 2.544
capstru
(0.035) (0.141) (0.042) (0.083) (0.637)
-39.7*** -126.508*** -17.537 -54.903** 30.939
_cons
(0.006) (0.007) (0.229) (0.020) (0.384)
R2 0.667 0.499 0.629 0.589 0.641
N 227 172 39 137 90
Table 2 – The equity incentive effect of Listed Companies in China
4.3. Analysis on the mechanism of stock incentive effect of Listed Companies

in China
Table 3 further shows the employee stock ownership plan effect of equity incentive
mechanism analysis, that equation (2) and equation (3) of the estimation results,
including (1) examines the effect of equity incentive is affected by ownership concentration
influence, (2) examines the equity incentive effect is affected by the macro economic
development impact. Column (3) - column (6) is a sub sample of the estimated results.
For the full sample enterprises, the main observation variables, the variable coefficient of
ESOP 4.329>0, at 1% significant level of probability of statistically significant, ESOP and
ownership concentration degree oc5 cross terms in the coefficient of variable -0.057<0,
and statistically significant at 1% probability level.
RISTI, N.º E5, 07/2016 21

dependent variable performance

sample sample privately operated state-owned
column (1) (2) (3) (4) (5) (6)
4.329*** -1.189 12.483*** -0.588 0.718 0.642
esop
(0.000) (0.276) (0.000) (0.937) (0.359) (0.487)
-1.679** -1.916*** -3.898* -5.151** -0.893 -0.879
lnincome
(0.016) (0.007) (0.075) (0.027) (0.225) (0.233)
10.810** 11.466** 22.747 22.727 8.698** 8.539**
capstru
(0.022) (0.019) (0.172) (0.214) (0.041) (0.047)
-0.057*** -0.148** -0.005
esop_oc5
(0.000) (0.010) (0.692)
0.233** 0.616 0.020
esop_economy
(0.015) (0.266) (0.812)
-35.077** -40.094*** -84.486* -114.613** -17.633 -17.420
_cons
(0.012) (0.005) (0.061) (0.017) (0.228) (0.234)
R2 0.734 0.692 0.589 0.518 0.630 0.630
N 227 227 172 172 39 39
Table 3 – Analysis on the mechanism of stock incentive effect of Listed Companies

The results illustrate the enterprise ownership, the more concentrated on corporate
equity incentive positive effects have a negative impact, which will weaken the incentive
effects of equity. The marginal impact of corporate performance on equity incentive
is =4.329-0.057*oc5, the total sample size of enterprise equity
concentration is 57.116, so =4.329-0.057*57.116=1.073>0. In column
(2), variable coefficient of ESOP -1.189<0, in 10% probability level is not statistically
significant, ESOP and macro economy cross variable coefficient for 0.233>0, at 1%
probability level statistically significant, the results illustrate external macroeconomic
growth is better, have a positive impact positive effect on corporate equity incentive,
enterprise performance of equity incentive marginal effects for
=-1.189+0.233*economy, for the full sample and the macro economic growth rate of
10.192, so =-1.189+0.233*10.192 =1.186>0.
The analysis results of the stock incentive effect mechanism under the sample of state
owned enterprises and private enterprises. For private enterprises, when the introduction
of equity incentive and ownership concentration degree of cross variable, the variable
coefficient of ESOP is 12.483, cross term variable coefficient is -0.148, both in the 5% level of
probability of statistically significant, the results with the enterprises in the whole sample
results is consistent, indicates that equity more concentrated will weaken the positive
effects of equity incentive. When the introduction of equity incentive and macroeconomic
time series cross variable, the variable coefficient of ESOP is -0.588, cross term variable
coefficient is 0.616, were not significant, but the sign of the coefficients with the sample
under the agreement, this paper still think macro economic growth rate rose high-energy
invisible to expand the positive effects of equity incentive. For a sample of state-owned
22 RISTI, N.º E5, 07/2016

enterprises, when the introduction of equity incentive and ownership concentration

degree of cross variables, variable coefficient of ESOP was 0.718, cross term variable
coefficient is -0.005, two were under 10% probability level is not statistically significant;
when the introduction of equity incentive and macroeconomic time series cross variable,
the variable coefficient of ESOP 0.642, cross term variable coefficient is 0.020, were not
significant, but can calculated and enterprise performance of corporate equity incentive
marginal impact is positive. For the enterprise operating income and asset liability ratio
of two control variables, the coefficient symbol and the significance of the 2 agreement
with the table, here is no longer a detailed statement.
These results indicate that Chinese listed company’s employee stock ownership
plan, there exists a positive incentive effect. The results and most scholars research
conclusion consistent, indicating that ESOP can really mobilize staff Holdings
(usually in the senior management incentive objects) working enthusiasm and ability,
by strengthening enterprise management, control cost and business strategy, to
improve the performance of enterprises. There is a positive incentive effect for private
enterprises and state-owned enterprises there is no incentive effect explanation is:
for state-owned enterprises, there are two main reasons, one is the actual situation of
the state-owned enterprise’s equity incentive object for general staff and management
personnel of enterprises, especially senior managers, most countries appointed, which
is the incentive object can not participate in the decision of enterprise management
level, the incentive effect is not obvious, on the other hand, the state-owned enterprise
performance appraisal index target by the relevant departments of the impact, in
operation, the strategy and the system is passive, the “hard constraints” makes the
incentive objects can not effectively play the level of management, equity incentive in
the state-owned enterprises limited in. Behavior and for the general private enterprises,
on the contrary, first of all enterprises operating mechanism is more flexible, enterprise
equity incentive targets are also the actual management about the business objectives,
the “right” (management rights, equity) consistent background makes incentive staff
goals and objectives, to fully mobilize the enthusiasm of the management on improving
the enthusiasm of enterprise performance.
As for the impact of external variables on the equity incentive effect, the positive effect of
ownership concentration will reduce the enterprise equity incentive, this paper considers
that there are two factors, one is the top five shareholders, including few business
objectives of equity incentive management object, so the five largest shareholder equity
ratio is higher. Business decision rights incentive target was smaller (major projects,
strategic restructuring still depends on the actual controller), this situation will lead to a
positive incentive effect is weakened; the two is more concentrated ownership, itself will
occupy the interests of small shareholders, the earnings management is more inclined
to the current distribution, but not as the next the capital, more empirical analysis also
indicates that the ownership concentration and corporate performance are negatively
correlated. Therefore, this endogeneity (incentive and performance may be endogenous
problem) has led to the weak effect of incentive. Macro-economic growth to improve
the equity incentive effect, apparently in macro-economic upturn, the positive effect of
micro enterprise performance, compared with normal (economic) growth rate. At this
time behavior of equity incentive for incentive object is concerned with the existence
RISTI, N.º E5, 07/2016 23

of an “invisible” welfare effects, namely enterprise performance ascending part of is by

external economic growth rate.
5. Conclusion
The agency behavior under the modern enterprise system due to the asymmetric
information between the owner and the manager in the two sides cannot be effectively
supervision and management, employee stock ownership equity incentive system to
gradually solve the problem. However, there are many researches about the effect of
employee stock ownership incentive on firm performance, but the conclusion is not
uniform. This paper chooses in the period 2006-2012 have the behavior of equity
incentive in China A-share listed companies, using paired samples t test and the cross-
sectional regression model, analysis of the impact on the effect of employee stock
ownership plan effect of equity incentive and external variables. The results show that:
the performance of Listed Companies in China before and after the equity incentive (net
profit growth rate) has significant growth. Employee stock ownership incentive can have
a positive effect on the performance of listed companies, but this effect is more significant
for private enterprises and small scale enterprises, while the effect is not obvious for
state-owned enterprises and large-scale enterprises. The positive effect of ownership
concentration on employee stock ownership incentive is affected by the impact, while
the macro economy has a positive effect on the positive effect of equity incentive.
Based on the above conclusions, this paper puts forward some suggestions on the
operation of employee stock options:
•• Develop a suitable to the enterprise own development incentive program,
currently listed companies incentive schemes and similar, no difference, are
inspired by the linked object and corporate performance. If the performance of
the company in a specified period can be completed the exercise. Although this
initiative is indeed able to mobilize the enthusiasm of executives, but whether
it is really suitable for the company’s development, more difficult to judge.
Enterprises from the incentive model and source of capital (stock), incentive
object, motivation scale, exercise price, exercise arrangement (period), the
exercise conditions (performance appraisal) full investigation and verification.
•• Enterprises should strengthen the system construction of employee stock
ownership incentive strategy. In determining the incentive plan, major
shareholders and decision-making layer to the equity incentive before the
implementation, and after exercise are clearly defined, especially for managers
(inspired by the top management object) to carry out effective supervision to
prevent incentive management personnel take damage to minority shareholders
and the company’s long-term growth behavior in order to achieve the goal
incentive, which can be from salary, performance evaluation and other aspects
of the regulation. On the other hand, you can consider the use of reverse
repurchase way to implement equity incentive, which can fully mobilize the
management level of existing shares executives.
•• Continue to improve the laws and regulations on the system of employee stock
ownership in china. The securities regulatory authorities should strengthen of
equity incentive of listed company information disclosure to crack down on
24 RISTI, N.º E5, 07/2016

insider information, standardize enterprise bonus system and strict supervision

on the behaviors of corporate executives, to develop more in line with the reality
of laws and regulations, and effectively protect the interests of small investors.
In the equity incentive, it is not only to give the company a certain right to
freedom, but also to allow the market, employees, and investors to effectively
supervise the three.
Acknowledgments
This study was financially supported by Humanity and Social Science Youth foundation
of Ministry of Education of China (Project NO: 15YJC630060); “Corporation Society
Responsibility” Academic Research Team of WTBU (XSTD2016001).
References
Informação, (16), 108–128.
Azadi, M., Shabani, A. (2014). Planning in feasible region by two-stage target-setting
DEA methods: An application in green supply chain management of public
transportation service providers. Transportation Research Part E: Logistics and
Transportation Review, (70), 324–338.
Chen, W., Jia, S. (2015). State owned holding, market competition and equity incentive
effect — An Empirical Study Based on propensity score matching method. Journal
of Zhejiang University, (9), 101–118.
Huang, G., Zhang, Y. (2009). An empirical analysis on the performance of State-
Owned Company’s employees’ Stock Ownership: Based on the sample data of 1302
companies. Economic science, (4). 86–94.
Kong, J., Xu, Y. (2015). An empirical study of ESOP incentive effect — Based on the
historical data of the private SMEs of Shaanxi province. Journal of Guangxi
University, (3), 68–74.
Shen, H. (2011). Empirical analysis of equity incentive effect of A stock market based on
event study .financial development research, (2), 73–77.
Wang, Y., Shen, X. (2015). Equity incentive, property rights and corporate performance,
equity incentive. Journal of Southeast University, (1), 71–79.
Wu, Z. (2014). The effect of board governance characteristics on equity incentive effect.
Journal of Guangdong University of Finance and economics, (6), 80–89.
RISTI, N.º E5, 07/2016 25

Study on the Image Supporting System and

Propagation Mechanism of Rural Tourism Destination
Li Long, Song Hui*
* songhuisuzhou@163.com
Suzhou University, Suzhou city, Anhui province, 234000, China

Pages: 26–36
Abstract: In this paper, the author analyzed the rural tourism brand building
and dissemination, which based on the basic concept and implementation path
from the tourism destination image building, image support system and image
propagation mechanism. For example in Anhui rural tourism destination,
through field investigation of rural tourism for empirical research, we found in the
northern Anhui rural tourism is still in its primary stage, image positioning is not
accurate enough, brand communication channel is single, also awareness is not
high, resulting in rural tourism destination brand competitiveness is not strong,
restricted the development of rural tourism seriously. Based on rural tourism
development present situation of northern Anhui, we proposed that strengthen
government guidance, the regional unified planning, combination of media, spread
consistent image, using emerging technologies, and rooted in brand culture, to
achieve sustainable development of Anhui rural tourism.
Keywords: Image supporting system, Tourism image shaping, Tourism
propagation mechanism, Tourism destination, 3D virtual interactive
1. Introduction
At present, the development of rural tourism in China quickly, especially since
countries in 2006 as “China rural tourism year” so far this time, development of
village tourism began in full swing. In between Eleven-Five and Twelve-Five, rural
tourism development meet opportunities, while there have been some issues during
rural tourism development, rural tourism destination image position is not accurate
enough, and brand competitiveness is not strong enough, serious constraints on the
development of rural tourism(Yong, 2002). To solve these problems, will provide new
opportunities for the development of rural tourism in our country during thirteen
five periods. Rural tourist destination image is the tourists’ first impression of rural
tourism will have a primary effect, halo effect, so tourism destination image plays an
important role, and rural tourism image dissemination effect affects the influence and
radiation radius of tourism brand. Good tourism image is designed and constructed
based on the elements of construction, from the actual, blindly packing speculation,
is not possible to establish a good image of tourism; but only the excellent design
and construction, no effective dissemination of the image, the tourists and the public
26 RISTI, N.º E5, 07/2016

cannot form a good impression, only the excellent design and construction of effective
communication together, to establish a good image of tourism. Rural tourist destination
image is composed of a variety of factors, its connotation is to continuously improve,
will experience from the image positioning is not clear to clear, and then to the process
of re-positioning the destination image(Chun, 2004; Freixo, 2014).
2015 China Netizen tourist and predetermined behavior survey report pointed out that
the number of China’s Internet users has reached 485 million. Among them, 34% of
Internet users over the past six months participated in tourism activities, 1.66 billion
(figure 1). This shows that China’s Internet users demand for travel and is very large;
many people want to be able to travel in a busy work.
Figure 1 – Internet users in the past six months whether to travel

Research has also shown that the future users within a year go out tourism greater
demand, overall Internet users, 47.2 percent of people travel demand, this ratio increase
13.2% than 2014 (Figure 2). Therefore, this part of the crowd has travel demand, but
because of various reasons, is unable to achieve the desire. And with the development of
economy, people’s demand for spiritual civilization will also be expanded, and the future
demand of Internet users is expected to increase.
Figure 2 – The expected of travel by internet users in the coming year

Rural tourism destination image first determine a clear theme, the theme of the selected need
to be based on a large number of basic theoretical research and field investigation(Mengxiang,
RISTI, N.º E5, 07/2016 27

Study on the Image Supporting System and Propagation Mechanism of Rural Tourism Destination
2006; Peiji, 2006). The establishment of rural tourism destination theme is the key to
shaping the image of the rural tourism destination, which has a direct relationship between
the formation of a rural tourist destination image and brand. Tourism theme image once
formed, will be in the minds of tourists have the general impression. Generally speaking,
the accumulation of the impression of the image is relatively stable.
Figure 3 – The formation of rural tourism destination image theme
2. Virtual reality technology and tourist image

2.1. Tourism experience
Tourism experience research began in twentieth Century sixty or seventy years, is the
core of tourism sociology and even modern tourism research, basic theory. At first, the
researchers gave the definition of the tourist’s experience, because of the nature of tourism
and experience to hold different views; there are differences in the understanding of
the tourism experience. In 1998, Richard C and prentice et al summarizes the five most
representative tourist experience research paradigm: in leisure studies in North America as
the representative of the level of the modes of experience; as a psychologist Zeke Center as
the representative of smooth cool “experience a paradigm; behavioral intention experience
paradigm; Mr. Cohen, for on behalf of the phenomenon of learning experience paradigm;
insider outsider paradigm. Subsequent to McCann Nair (J) to represent the presence
of authenticity experience research paradigm and symbol science from the perspective of
experience research paradigm from behind, the understanding of tourism experience from
the tendency to understand the experience to understand the essence of tourism (Wei,
2005). Pace of tourism study on modern leisure tourism advantages and disadvantages of
heated debate strode, keen to explore the problem of “authenticity” of tourism experience.
Finally, Cohen to learn from the perspective of the tourism experience research paradigm
unifies the differences become tourist experience research important theory and scope of
the research type and lead the study of Tourism experience to the demand of tourists,
motivation types and quality of tourism experience research.
Boorstin in 1964 with the old tourist seeking knowledge of the view of the modern public
to complain about the pursuit of artificial, hypocritical travel experience proposed the
28 RISTI, N.º E5, 07/2016

modern mass tourism experience is just a popular consumer behavior. 1975 Turner
and Ash also pointed out that the modern mass tourism is a departure from the normal
behavior, is a disease of the times. MacCannell used “Modernity” to explain “authenticity”,
argues that the modern people to tourism in the pursuit of truth, suffered a lot of criticism
to escape the modernity, which tourism sociologists in 1979 Cohen in the published the
travel experience of phenomenology of the additional monarch paths to phenomenon
from the perspective that, due to the different people the ultimate meaning and world
view is different, different experiences for different tourists and different community
has different meanings, so different people need different experience. Tourist experience
is defined as individuals with a variety of center; depending in person is attached to a
“center” tourism experience to meet the travel from the pursuit of pleasure to search
for the meaning of demand. Cohen, this travel experience research paradigm from the
perspective of phenomenology, the travel experience “authenticity” of the differences
were unified, and foreign scholars have been widely recognized and adopted (long, 2011).
On this basis, the researchers began to study the needs of tourists, motivation, experience
quality, experience content and Realization of the path. At first, Pearce inspired
by Maslow’s theory of hierarchy of needs, the needs of the tourists are divided into
relaxation, stimulation, relationship, self-esteem and development and self-realization.
Later, Ryan Chris will travel demand summed up as intellectual, social, ability, avoid
stimulation. The research of tourism demand lays the foundation for the quantitative
and Empirical Study of tourism experience. American scholar Mcintosh first proposed
four types of tourism motivation, including: physical health, culture, communication,
status and prestige. Japanese scholars use factor analysis to the tourism motivation:
mood, physical, spiritual, economic. Hui (2008) put forward eight categories of tourism
motivation: landscape tour, fitness tourism, culture, arts, tourism, urban shopping and
sightseeing, entertainment, Cuba tourism, community tourism, food tourism and explore
tourism of urban life. In addition there are scholars summed up the tourist motivation of
the backpack tourists, including: escape, relax, knowledge, search, experience and social
aspects of the five. There are scholars to study self-help travel motivation include: the
pursuit of freedom, to get a sense of challenge, the pursuit of a unique sense of self.
Other scholars found that the five impact factors of Taiwan tourists’ travel experience
are: relaxation and fun, cultural entertainment, personality identity, historical memory,
escape. Study on the quality of tourism experience. The theme of the study is to explore
the tourists experience to meet the formation model, experience the quality of the factors
and to experience the quality of the evaluation and enhance the experience of quality
strategy research. According to the research method, it can be divided into qualitative
research and quantitative research, in which the latter is the main research method.
2.2. Virtual reality technology

The intervention of the virtual reality technology to virtual tourism is through computer
simulation reality of tourist attractions to generate realistic audio-visual collection is one of
the three-dimensional space, according to inherited the historical and cultural background,
through 3D virtual interactive technology reproduction space and historical monuments,
let has been quiet in monuments of historical culture or story can be reproduced in the
appropriate context, visitors can even with its own set of role and the historical space
characters dialogue, to learn more about what had happened in a particular historical period.
Let the tourists “real” participation in the cultural context of the similar historical context,
RISTI, N.º E5, 07/2016 29

the more true feeling of historical and cultural. Through the immersive three-dimensional
experience of the digital scenic area, get the intuitive, real experience, so that the relatively easy
access to relatively comprehensive historical and cultural information and the “immersion”
experience. The Louvre is built on the Internet “virtual museum” the collection of hundreds
of thousands of precious exhibits posted online for public access, and with full details of the
depth to the text explanation and pictures. But in the virtual online museum, the public is
through pictures and text of the panoramic view. But its shortcoming is show the lack of
more realistic immersive experience, such as when visitors in viewing a painting, only to see
the painting pictures and the following text explanation, there is no feelings from the direct
experience of the cultural connotation of historical relics, as shown in figure 4.
Figure 4 – The Virtual Museum of Louvre

The Mogao Grottoes of Dunhuang due to the harsh natural environment and traffic
inconvenience, so many Dunhuang art lovers to witness the Dunhuang art treasures,
2003, President of the Dunhuang Academy submitted the tourist service center of
Dunhuang Mogao Grottoes construction proposal “proposal, the main content of this
proposal is: request to digital technology and computer technology and other modern
means, the Mogao Grottoes of Dunhuang digital results display to tourists.
Figure 5 – The Virtual Museum of Louvre
30 RISTI, N.º E5, 07/2016

The virtual tour breakthrough time and space

constraints, the contradiction between the
Dunhuang Grottoes in the preservation and
tourism development have been alleviated,
and provide a new perspective on visit to
tourists and Dunhuang researchers, as shown
in figure 5 and figure 6. At the same time, due
to the reduction of the actual contact time
between the cave and tourists, will be conducive
to the stability and long-term preservation of
the caves. Because this kind of tourism is in
the computer virtual of, many scenes not seen
in the Dunhuang travel can also be displayed,
currently on display in the main content
including Dunhuang Mogao Grottoes location
roaming, Dunhuang Cave virtual excavation
process display, virtual restoration of Shi Kuxu
intends to roam, the real grottoes of virtual
roaming and creation of computer aided cave
virtual roaming and other.Virtual the Imperial
Figure 6 – Virtual tourism of the
Palace is China and Asia, as shown in figure 7,
recovered grotto (Mogao Gave205)
is the first on the Internet to show the world’s
historical and cultural heritage of the virtual
world, but also a network experience museum.
Visitors can choose their own role, and give it to change clothes, dress up and so on.
Visitors with the mouse, you will be able to follow the role of in-depth understanding of
all the buildings in the Imperial Palace and its characteristics.
Figure 7 – Virtual the Imperial Palace
3. Village tourism image systems

Rural tourist destination image by image positioning and tourism slogan designing,
shaping a distinctive destination image, but these are not enough, they also need to
RISTI, N.º E5, 07/2016 31

improve the support system. Supporting system by mind identity system (MIS), visual
identity system (VIS), and behavior identifies system (BIS) and so on.
3.1. Mind identification system

The mind of rural tourism is the core and soul of rural tourism image design, a deep
analysis on the basis of the concept of rural tourism must be in a wide range of design
and development of rural tourism products. Because of the rural tourism is a unique
living environment, production patterns, life style and idyllic scenery to attract tourists,
it reflects a return to nature and artistic conception of the unity of heaven and man,
also in line with the rise great trend of “green tourism”, “eco-tourism” in recent years.
Therefore, I believe that based on ecological protection and sustainable development of
the rural cultural heritage tourism is a rural tourism image design of common conceptual
basis. Of course, different rural tourism and tourism products, due to the geographical
position, natural resources and market demands and other reasons, its conceptual basis
should be different, focused, to reflect their personalities and characteristics, which
requires on tourist destination image design to concrete analysis of concrete problems.
3.2. Behavior identification system

Identification of rural tourism is tourism image dynamic recognition of forms, is
the activity of the rural tourism resources, and it is the practice of philosophy and the
creation of guidelines for tourism and culture. At the time of rural tourist destination
image design, focus on images of local residents to create and design of tourist image
of enterprises, especially in China.
Figure 8 – Tourism image recognition systems
32 RISTI, N.º E5, 07/2016

3.3. Visual identification system

According to relevant statistics, perceived as external information, 83% through the
Visual channel to reach the minds of people, meaning that vision is the acceptance of
external information of the main channel. Visual recognition systems can all be visible
the external visual symbols to convey the operating concept with intelligence information
in the CIS is the most direct and effective way to establish corporate reputation and
shape the corporate image. Visual identity design for rural tourism usually includes
visual identity design and Visual landscape design. At present, new media technology,
widely used in various fields, but is still at an early stage of tourism development in rural
areas, rarely used new media technology. Thus at the time of rural tourist destination
image design,to make full use of new media technology spread rural tourism destination
image, for tourists to enjoy the visual.
In addition to the above three identification systems,there are some scholars and
auditory identification system (HIS), customs identification system (FIS). Auditory
images generally include the destination language, folk songs,local opera, background
music and theme songs and religious music. Customs identification means in rural
tourist destinations only and representative image to become the region’s program
or activity. Thus, when the rural tourist destination image, also want to focus on the
auditory system and design style identification system. Comprehensive support system
is a rural tourist destination image design’s strong backing, is the internal factors of
sustainable development of rural tourism destinations, and it is also essential for rural
tourism brand of key components.
4. The transmission system of rural tourism destinations image

Tourism communication mechanism means how to spread during the trip, all aspects
of the tourism communication link together organically to form travel a reasonable
flow of information channels. Rural tourist destination image design approaches for it
to spread in the future, only the rural tourism destination image out to make tourists
feel, to make tourists travel to this desire. Different routes of transmission will have a
different effect, spreading the effects is related to the tourism image and brand in the
minds of the impression level.
4.1. Dissemination channels and the negative effects of eliminating

From the interface point of view, tourism communication through self, people, the
general public and the information highway and multimedia communications. Rural
tourism in the different media to disseminate results will vary, so when you select the
media, to specific conditions, for different tourist market select targeted media. Image
media including mass media, interpersonal media, outdoor media, real media, tourism
image media classification and characteristics are described in detail.
From information dissemination pattern of de Fleury, we can see that rural tourism
destination by a specific transmitter that transmits a variety of communication channels
to mass media, people perceive the image of rural tourism through various mass media,
and feedback on this information. In the process of information transfer will be in the
RISTI, N.º E5, 07/2016 33

Figure 9 – De Fleury model
“noise” effects of tourist image in the process, which will suffer the effects of negative
events about rural tourism, damaging tourism destination image in people’s perceptions.
Thus, in rural tourism image in the process of spreading propaganda, should pay
attention to reduce or eliminate the negative impact on rural tourism destination image.
4.2. Empirical study on effect perceptions

Investigating present situation of rural tourism development in Northern Anhui,
stratified random sampling survey, the questionnaire analysis by SPSS17.0 software, we
find the rural tourism destination of tourists is the surrounding area, this mainly due
to rural tourism attraction radius and the spread effect of restricting the scope of the
destination image. Tourists’ perception of rural tourism in northern region is mainly
based on network and the main newspapers magazines, because single transmission of
the scenic routes, there is no targeted market segments spread enough in the network
media and other reasons, caused by propagation effects of tourism destination image is
not ideal. Thus, in rural tourism destination image communication, we should segment
tourist market for different markets, select the appropriate media for dissemination
of tourism image. Meanwhile, using a variety of modes of transmission are combined
communication and in the dissemination of note also that in the process of creating a
unified tourism rural tourism destination image.
Factor Frequency Percent Valid Percent Cumulative Percent

Travel agency 21 14.0 14.0 14.0
TV 6 4.0 4.0 18.0
Network 66 44.0 44.0 62.0
Relatives and friends 15 10.0 10.0 72.0
Newspapers and magazines 42 28.0 28.0 100.0
Total 150 100.0 100.0
Table 1 – Tourists’ perception of Rural Tourism in North Anhui
34 RISTI, N.º E5, 07/2016

5. Conclusion and recommendations

Create a rural tourism destination image plays an important role in rural tourism
planning and development, which related to a tourism development success, so we
must pay attention to the design and building of rural tourism image. However, due
to the development of rural tourism in China not long ago, research is weak, low level
of awareness, its development plans tend to be resource-oriented development or
market-oriented development, ignoring the tourist image. In the rural tourism business
process is also a blind pursuit of short-term economic benefits, ignoring the protection
of the rural overall image and rural tourism environment and traditional culture. This
is unfavorable to the long-term sustainable development of rural tourism. In the early
stage of the rural tourism development, government should pay attention to the guiding
role, the government to guide the rural tourism destination image, the long-term goal of
unified planning of rural tourism development. In the process of rural tourist destination
image communication, we must according to the specific situation to carry out targeted
communication. Dissemination of the image of rural tourism should focus on the target
market of tourists and potential tourists, using the appropriate route of transmission for
effective dissemination, all means of communication should be spread around strategic
goals to launch, through a combination of media, passed on to the unified rural tourism
image of tourists.
Along with economic development, social progress and people’s conception of the
renewal, rural tourism has become the new focus of residents’ consumption in China,
especially urban resident. Tourism image and communication, to some extent, means that
the expansion of economic efficiency and market, it is the key element to attract tourists.
Therefore, in order to create a distinctive rural tourism image, keeps the rural tourism
destination attraction for a long time, it should be based on the country’s own characteristics
and development trend of tourism consumption, establish their own unique image, as
well as strong advocacy marketing campaign to spread the rural tourist destination image,
and rely on village tourism image to attract tourists, promote the development of rural
tourism. In parallel with the development of rural tourism, we should stick to the rural
ecological protection based on sustainable development of tourism and rural cultural
heritage concept. In the thirteen five period, rural tourism will be developed to the depth
direction, we believe that rural tourism will take a scientific concept as the foundation,
create a perfect rural tourism destination image, create their own brand, get out of the
road of sustainable development with its own characteristics.
Acknowledgments
Song Hui is corresponding author. This paper is supported by Key project of Humanities
and social sciences of Anhui Provincial Department of Education “The study area
tourism development under the background of urbanization in Wanbei Coal Mining
subsidence” (No: SK2016A1003); the Humanities and social sciences research key
project of Anhui College “Evaluation and utilization of wetland tourism resources in
Northern Anhui”(SK2015A565).
RISTI, N.º E5, 07/2016 35

References
Chun, L. (2004). Tourism Psychology (Second Edition), Beijing: China higher education
press, 20–32.
Freixo, J., & Rocha, Á. (2014). Arquitetura de Informação de Suporte à Gestão da
Qualidade em Unidades Hospitalares. RISTI-Revista Ibérica de Sistemas e
Tecnologias de Informação, (14), 1–15.
Hui, H., Jian, X. (2008). Modern tourism communication. Hunan science and
technology press, 22, 22–25.
long, L. (2011). The construction of brand communication path in the context of new
media tourism destination. Economic Research Guide, 5, 30–35.
Mengxiang, Y. (2006). The visual recognition system into tourism related research
(VIS). Journal of Guilin Institute of Tourism, 10, 23–30.
Peiji, S., Xianfeng, L. (2006). Study on tourism image communication. Journal of
Southwest University(Social Sciences), 10, 8–10.
Wei, X. (2005). Rural tourism destination image design based on TIS. Ecological
Economy, 6, 23–26.
Yong, M., Xi, L. (2002). Tourism planning and development. Beijing: higher education
press, 12, 4–8.
36 RISTI, N.º E5, 07/2016

Countermeasures and Support Environment for the

Development of the Non-governmental Economy in
Suzhou, China
TianXueying*, NiePanke
* lucy.txy@163.com
School of Business, Suzhou University of Science and Technology, Suzhou215009, Jiangsu, China
Pages: 37–49
Abstract: As required by the policy of “New Normal”, more attentions need to

be paid to the quality and efficiency of the economic development. Therefore, it
is important for the non-governmental economy in Suzhou, an important pillar
experiencing rapid development, to learn how to recognize and adapt to the
new normal, and deal with the accompanied new opportunities and challenges.
The purpose is to motivate the vitality and promote the transformation and
upgrading of the non-governmental economy, so as to enhance the comprehensive
competitiveness, which is the key for the sustainable economic growth. By
analyzing the overall scale, industrial structure, social contribution, scientific and
technological level of the non-governmental economy in Suzhou, this research
revealed the problems facing this economy. On this basis, this research explored
corresponding countermeasures in terms of promoting the adjustment of the
industrial structures and the characteristic development. Meanwhile, it is also
suggested to improve the product quality, brand building, the investment on
research and development, independent innovation ability, the energy efficiency,
and comprehensive utilization of resources. Meanwhile, some policy supports were
proposed based on the talent, financial, and government environments, so as to
provide a scientific basis for the decision marking of the governments.
Keywords: Non-governmental economy, problem analysis, support environment,
countermeasures
1. Introduction
The president, Xi Jinping, systematically elaborated the meaning of the new normal in
China’s economy in the Asia Pacific Economic Cooperation (APEC) held in 2014. Since
then, how to adapt to the new normal and grasp new opportunities, so as to realize new
growth points has been a historical problem facing many enterprises, especially for the
non-governmental enterprises. The non-governmental economy in Suzhou has witnessed
vigorous development and presented a favorable growth in both the quantity and quality
since 2009. Meanwhile, the enterprise scale and operating income have increased
substantially, and the proportion of the paid taxes has risen year by year (Pan, 2006;
Freixo, 2014). In addition, remarkable growth has been found in the private investment,
and with the increasingly advanced scientific and technological level of the enterprises,
RISTI, N.º E5, 07/2016 37

Countermeasures and Support Environment for the Development of the Non-governmental Economy in Suzhou, China
the enterprises and entrepreneurs of high quality have emerged continuously. The
rising of the non-governmental economy in Suzhou has laid a solid foundation for the
scientific development of the economic society in Suzhou. As required in “New Normal”,
more emphasis is required to be focused on the quality and efficiency of the economy
(Jiang,2009; Zhang, 2011). Meanwhile, as an important pillar, the rapidly developed
non-governmental economy in Suzhou has to learn to recognize and adapt to the new
normal, as well as grasp the new opportunities and address the challenges posed by the
new normal. The purpose is to motivate the vitality and promote the transformation
and upgrading of the non-governmental economy, so as to enhance the comprehensive
competitiveness. This is the key to maintain the sustainable economic growth, and also
an urgent research problem (Chen, 2009; Lu, 2013). There have been disputes regarding
the understanding and the definition of the scope of the non-governmental economy.
However, most scholars held that non-governmental economy is not equal to the private
economy, and apart from the private economy, non-governmental economy also includes
the private economy contained in the mixed economy. According to the definition of the
non-governmental economy in the Report on the Development of the Non-governmental
Economy in China 2003 (the Report 2003 for short hereinafter), on the one hand, non-
governmental economy is a general designation to the multi ownership economy which
is exclusive of the state owned and state holding enterprises. It includes the individual
business households, private enterprises, collective enterprises, enterprises invested by
Hong Kong, Macao, Taiwan and foreign countries. On the other hand, non-governmental
economy does not include the enterprises invested by Hong Kong, Macao, Taiwan and
foreign countries in a narrow sense. The non-governmental economy appearing in this
report generally refers to the non-governmental economy in a narrow sense. Based on
the narrow definition of the non-governmental economy in the Report 2003, the private
enterprises and individual workers were incorporated into the non-governmental
economy in this research. By doing so, this research aimed to explore the development
perspective of the non-governmental economy in Suzhou under the background of the
new normal, so as to reveal the problem facing the economy. Based on this, this research
explored corresponding countermeasures so as to provide scientific evidence for the
government’s decision-making (Esteban, 1994; Qian, 2013).
2. The Current Situation and Problem of the Non-governmental

Economy in Suzhou
2.1. The Overall Scale
The overall scale of the non-governmental economy in Suzhou is great, and has been
increasing year by year. As shown in the latest statistical yearbook of Suzhou in 2014,
the number of the registered private enterprises and individual business households was
705,404 in 2013, in which there were 255,722 registered private enterprises, increasing
by 11.37% compared with the 229,618 in 2012. Meanwhile, the number of the individual
business households was 449,682, increasing by 8.35% compared with the 415,023
in 2012. The total number of the employees in the non-governmental economy was
4,438,447 in 2013, which presented an increase of 14.6% compared with that in 2012.
Thereinto, the numbers of the employees in the private enterprises and the individual
business households were 3,539,530 and 898,917 respectively, which increased by
15.37% and 11.51% separately compared with those in 2012. The total amount of the
38 RISTI, N.º E5, 07/2016

registered capitals in the non-governmental economy was 995.41876 billion yuan in

2013, presenting an increase of 14.5% compared with that in 2012. Thereinto, the total
amount of the registered capitals in private enterprises and the registered capitals per
household were 966.20638 billion yuan and 3.7783 million yuan separately, which
increased by 14.15% and 2.5% respectively compared with those in 2012, as shown in
Table 1. In addition, as demonstrated in the statistical bulletin regarding the national
economy and social development of Suzhou in 2014, the number of the registered private
enterprises and individual business households in Suzhou was 785,200 by the end of
2014, increasing by 11.3% compared with that of the last year. Thereinto, the numbers
of the private enterprises and individual business households were 295,600 and 489,600
respectively, which presented an increase of 15.6% and 8.9% separately compared with
those of 2013. Meanwhile, the amount of the registered capitals in private enterprises and
individual business households was 1,183.02 billion yuan, growing by 18.8% compared
with that in 2013. The total amount of the private investment was 332.18 billion yuan,
presenting an increase of 1.8% compared with that in 2013, and accounted for 53.3% of
the total investment in the fixed assets.
increased
Main indexes 2013 2012
proportion
The total number of the enterprises 255722 229618 11.37%
The number of the employees 3539530 3067910 15.37%
Private Registered capital (ten thousand
enterprises 96620638 84646807 14.15%
yuan)
Registered capital per household
377.83 368.64 2.5%
(104yuan)
The total number of households 449682 415023 8.35%
Individual
business The number of the employees 898917 806147 11.51%
households
Registered capital (104yuan) 2921238 2514328 16.18%
Table 1 – The overall situation of the non-governmental economy in Suzhou
2.2. The Industrial Structure

The industrial structures in the non-governmental economy of Suzhou are generally
in the following order: tertiary industry> secondary industry> primary industry from
the perspective of the numbers of the enterprises and employees, and the proportion
of the registered capitals. Meanwhile, tertiary industry accounts for a great proportion
in the non-governmental economy. As for the primary industry, except the registered
capital of the individual business households accounts for 1.4%, other indexes of the
individual business households and all the indexes of the private enterprises are all
less than 1%. Meanwhile, tertiary industry plays an absolute dominance in terms of the
numbers of the enterprises and employees, as well as the registered capitals. Thereinto,
as for the private enterprises, the numbers of the enterprises and employees, and the
RISTI, N.º E5, 07/2016 39

registered capitals are 156,894, 1,458,883 and 612,146.51 million yuan separately, which
account for the proportions of 61.3%, 41% and 63.3% respectively. The numbers of the
individual workers, related employees and registered capitals are 387,677, 682,378 and
2 2825.38 million yuan respectively, occupying the proportions of 86.2%, 75.9% and
78.1% separately, as shown in Table 2.
According to the analysis on the financial situations of the non-governmental industrial
enterprises above the designated size in 2013, the sectors where the main output
value and profit of the non-governmental industries above the designated size are
concentrated can be acquired. These sectors include the manufacturing industries of
the electrical machines, equipments, steel wire products and universal equipments,
the smelting and rolling processing industry of ferrous metals, manufacturing industry
of chemical fibers, textile and clothing. Obviously, except the textile and clothing
industries, all of the aforementioned industries belong to the heavy industry, and are
important pillars for the non-governmental enterprises. However, although these
industries have created primary economic contribution, they present great pollution
and energy consumption. In 2013, the tertiary industry in the non-governmental
economy of Suzhou mainly includes the traditional service industries such as wholesale
and retail businesses, leasing and business service industries, neighborhood service and
other service industries, accommodation and catering industries and so on. In contrast,
the strategic emerging industries such as new energy, new materials, biotechnology,
and new medicines take a small proportion. Meanwhile, they generally exhibit the
drawbacks including small scale and capital, single product structure, and low added
value and production efficiency.
Secondary industry Primary industry Tertiary industry

Main indexes
Quantity Proportion Quantity Proportion Quantity Proportion
enterprises
1501 0.6% 97327 38.1% 156894 61.3%
number
The
Private employees 14316 0.6% 2066331 58.4% 1458883 41%
enterprises number
Registered
capital 342509 0.4% 35063478 36.3% 61214651 63.3%
(104yuan)
households
1649 0.4% 60356 13.4% 387677 86.2%
number
employees
Individual 4119 0.5% 212420 23.6% 682378 75.9%
number
workers
Registered
capital 39785 1.4% 598915 20.5% 2282538 78.1%
(104yuan)
Table 2 – The industrial structure of the non-governmental economy in Suzhou in 2013
40 RISTI, N.º E5, 07/2016

2.3. The Social Contribution

The development of the non-governmental economy has offered considerable
job opportunities for the society. On the one hand, an increasing number of rural
populations flow into the urban areas, which promotes the urbanization process. On the
other hand, the workers laid off from the state-owned enterprises can be reemployed
by the non-governmental enterprises, which accelerate the reform of the state-owned
enterprises and promote the stability of the society. According to the latest statistical
yearbook of Suzhou in 2014, the total number of the employees working for the non-
governmental economy was 4,438,447 in 2013, accounting for 63.8% of those in all
economy types. This is the greatest contribution that the non-governmental economy
in Suzhou makes to the society. In addition, according to the statistics on the financial
situation of the above-scale non-governmental industries, it is discovered that there
were 5,910 enterprises in the above-scale non-governmental industries, presenting
an increase of 6% compared with that in 2012. Meanwhile, the gross industrial output
value, total assets, major business income and total profits created by these enterprises
were 1,009,750.34, 933,807.98, 1,004,889.69 and 34,167.3 million yuan respectively,
which increase by 10.2%, 11.3%, 8.8% and 7.6% separately compared with those in 2012.
The specific details are illustrated in Table 3. It is obtained that the non-governmental
economy in Suzhou is experiencing rapid development.
Main indexes 2013 2012

The overall condition of the non-governmental economy
The number of the employees 4438447 3874057
The proportion of the employees working for the non- 63.8 55.8
governmental economy in the total employees of the city (%)
The financial condition of the above-scale non-governmental industries
The number of the enterprises 5910 5577
Gross industrial output value (104 yuan) 100975034 91636331
Total assets (104 yuan) 93380798 83902432
Major business income (104 yuan) 100488969 92348541
Total profit(104 yuan) 3416730 3255194
Total profit and taxes (104 yuan) 5722267 5318733
Table 3 – The contribution of the non-governmental economy to Suzhou
2.4. The Scientific and Technological Level

As for the expenditures spent on the scientific research in the large and medium-
sized industrial enterprises in 2013, 1,554.83 million yuan was spent on the technical
transformation of the private enterprises. This was merely inferior to that of the limited
liability companies, and accounts for 27.8% of the total expenditures. The expenditure
RISTI, N.º E5, 07/2016 41

spent on the technology introduction was the minimum for being 56.41 million yuan,
which ranked the last place and accounted for a proportion of 3.2%. The expenditure
used to purchase domestic technology was little for being 6,4.31 million yuan, ranking
the second from the end and occupying 3.3% of the total expenditures. Besides, 3,471.38
million yuan was spent on the internal research and development, (R&D) which ranked
the fourth place and occupied 14.8% of the total expenditures. Compared with 2012,
all expenditures were increased except the expenditure used for purchasing domestic
technology, as demonstrated in Table 4.
Expenditure
Expenditure Expenditure spent on
Total internal
spent on spent on the purchasing
Item expenditure
technical technology the
spent on R&D
transformation introduction domestic
technology
Total amount 558532 170933 198791 2342437
Type of businesses
State-owned enterprise 3517 - - 6461
Collective Enterprise - - - 976
Cooperative stock corporation - - - 1722
Associated enterprise - - - 130
Limited liability company 169081 76222 167432 377927
Limited company 59480 6299 12417 162970
Private enterprise 155483 5641 6431 347138
Other enterprises - - - 2252
Enterprise invested by Hong 63050 41345 3821 437636
Kong, Macao and Taiwan
Foreign enterprise 107921 41427 8689 1005225
Table 4 – The expenditures spent on the technical transformation and introduction, absorption,
research and development in the large and medium-sized industrial enterprises in 2013
(104yuan)
The professional level of the technical personnels plays an important part in the research,
development and innovation of the enterprises. Table 5 displays the distribution
of the technical personnels in the enterprises of different economic types at the end
of 2013. It can be seen that the number of the technical personnels was 499,327 in
non-governmental enterprises, which was merely lower than that in the enterprises
funded by foreign countries, Hang Kong, Macao and Taiwan. Meanwhile, this number
accounted for 40.4% of that in the enterprises of various economic types. The numbers
of the personnels with senior and middle titles, postgraduates and undergraduates were
16,517, 110,660, 15,911 and 186,810 respectively, which occupied 3.3%, 22.2%, 3.2% and
42 RISTI, N.º E5, 07/2016

37.4% of the total personnels separately. These numbers all presented different degrees
of growth compared with those in 2012.
In general, non-governmental enterprises pay more attention to the technological
innovation and transformation. By the end of 2013, 505 non-governmental enterprises
have been engaged in the scientific and technological activities, accounting for 23.9% of
all enterprises. Among these non-governmental enterprises, 70.3% (355) participated
in R&D activities. With the enhanced ability of the scientific innovation and accelerated
scientific innovation, the number of the independent intellectual properties has been
increased. In 2013, the number of the patent applications was 4317 in private enterprises,
accounting for 27.3% of the total patent applications. Meanwhile, the number of the
invention patents in private enterprises was 3089, which showed an increment of 34
compared with that in the foreign-funded enterprises and ranked the first place among
all kinds of enterprises. However, few funds have been input into the innovation process.
Total Senior Middle Junior

Type Postgraduate Undergraduate
amount title title college
Total amount 1234777 55800 308325 55941 470340 500440
Public institution
owned by the 110505 16370 57771 6493 70570 26311
whole people
State-owned
9296 730 2433 721 4875 2847
enterprise
Enterprise
subordinated to
31210 3695 7141 5287 16448 6887
eachministry or
the province
Enterprise by
foreign countries,
Hang Kong, 584439 18488 130320 27529 191637 264508
Macao and
Taiwan
Non-
governmental 499327 16517 110660 15911 186810 199887
enterprise
Table 5 – The technical personnels in the enterprises with different economic

types at the end of 2013
In 2013, the expenditure spent on the technical transformation was 1,554.83 million
yuan in private enterprises, occupying 27.8% of the total expenditures. However, the
expenditure spent on the technology introduction was the minimum for being 56.41
million yuan, accounting for 3.2% of the total expenditures. The expenditure used for
purchasing domestic technology was 64.31 million yuan, merely with a proportion of
3.3%, which was lower than that in 2012. The internal expenditure spent on R&D was
3,471.38 million yuan, which occupied 14.8% of the total expenditures. It is inevitable
RISTI, N.º E5, 07/2016 43

that the insufficient input on innovation will affect the innovation ability and finally
impede the transformation and development of the non-governmental economy.
Meanwhile, there are absence of management talents and technical personnels in
the non-governmental enterprises. In 2013, the number of the technical personnels
in the non-governmental enterprises was 499,327, taking a proportion of 40.4%.
However, in these technical personnels, those with senior title accounted for 3.3%,
which was far lower than those in the public institutions owned by the whole people
(14.8%) and state-owned enterprises (7.9%). Meanwhile, the postgraduates in the non-
governmental enterprises took a proportion of 3.2%, which was lower than those in the
public institutions owned by the whole people (5.9%), state-owned enterprises (7.8%)
and enterprises funded by foreign countries, Hong Kong, Macao and Taiwan (4.7%). It
can be clearly seen that non-governmental enterprises have small proportions of the
technical personnels with high technical level and educational background.
3. The Countermeasures for Promoting the Healthy and Rapid

Development of the Non-governmental Economy in Suzhou
3.1. Promoting the Adjustment of the Industrial Structures and the
Characteristic Development of the Non-governmental Economy
Firstly, strategic emerging industries are supposed to be developed energetically. While
maintaining the competitive advantages of the traditional industries, the developing
plan of multiplying the emerging industries has to be implemented perseverely. That
is to say, non-governmental enterprises are encouraged to develop in the following
strategic emerging fields including new energy, new materials, new medicines, energy
conservation and environmental protection, biotechnology and internet of things.
Meanwhile, key support projects are set in these fields, accompanying with corresponding
preferences related to financing, taxes and scientific research, so as to gradually promote
the proportion of the emerging industries in the non-governmental economy of Suzhou.
Secondly, modern service industries are expected to be developed rapidly. It is suggested
to fully utilize the policy measures made by the governments of various levels for
promoting the development of the service industries and enhance the support degree
to the non-governmental service enterprises. Meanwhile, the policies proposed by
the municipal government to develop the modern service industry by separating the
manufacturing industries and the enterprises are required to be put into practice. By
doing so, the producer services such as scientific research, modern logistics, trade
marketing, specialized supply, industrial design, energy conservation service, and so on
can be promoted.
Thirdly, the characteristic development of the non-governmental economy needs to be
promoted. More attention is expected to be paid on small and medium-sized technology-
based enterprises, and the small and medium-sized enterprises with rapid growth, which
are characterized by great growth space, strong specialty characteristics and high technology
content. Meanwhile, by applying the specialty, refinement and unique as the objects,
these enterprises are subdivided and the characteristics are extracted on the precondition
of satisfying the common requirements. Afterwards, the all-round services including
information communication, product development, quality detection, management
consultation and technology introduction are pointedly provided in these enterprises.
44 RISTI, N.º E5, 07/2016

3.2. Striving to Improve the Product Quality and Enhancing the Level of
Brand Building
Firstly, policy guidance is utilized and enterprises are encouraged to produce the products
with high technical content, great market capacity and high added value. Meanwhile, top
priority requires to be given on the quality of the products by energetically implementing
the overall quality control system and standards. Besides, the quality of products needs to
be evaluated by the quality supervision organizations of various levels. In addition, non-
governmental enterprises are encouraged to strive for the various high-quality awards.
The enterprises with high quality and excellent brands are supposed to be awarded, and
advertised as the benchmark, model and star enterprises through vigorous propaganda,
so as to stimulate other enterprises to imitate and learn.
Secondly, it is suggested to implement the relevant rules of intellectual property
management issued by the state, Jiangsu province and Suzhou, so as to comprehensively
promote the number and quality of the patent applications in the non-governmental
economy. Meanwhile, attentions are supposed to be paid to the registration, protection
and management of the intellectual properties such as trademarks and brands, so as
to develop a group of model and competitive enterprises with intellectual property.
Enterprises are encouraged to implement the brand and famous brand strategies,
and thus enhance the brand consciousness and quality consciousness. In addition, it
is necessary to comprehensively promote the establishment of the quality system, so
as to adopt the idea of producing high quality products, and building famous brands
and favorable image for the non-governmental enterprises as objectives. Moreover, the
registration, application, management and protection levels of the trademarks require
to be further improved.
3.3. Increasing the Research Input and Enhancing the Capability of

Independent Innovation
Firstly, the various policies for supporting the technical innovation are supposed to be
implemented. These policies include the pre-tax super-deduction of research expenses,
favorable income tax for high-tech enterprises and technically advanced service
outsourcing enterprises, and rewarding the enterprises that establish the research
institutions and undertake the key science and technology projects of the state. Non-
governmental enterprises are encouraged to carry out the research and development on
the products and technologies based on the innovation. Meanwhile, enterprises should
appeal innovative talents and increase the research and development input by applying
the innovation and excellence as the objectives, so as to enhance the comprehensive
competitiveness. Meanwhile, non-governmental enterprises should employ, utilize
and retain the various management and technical talents according to their own
characteristics. The government should assist the non-governmental enterprises in
enhancing the intensity of talent introduction by providing service platforms. Various
policies and measures are utilized to guide and help the non-governmental enterprises
to promote the creation of the intellectual property through independent development
and university-industry cooperation. This can promote the transformation and
industrialization of the scientific and technological achievements, and thus enhance the
innovation capacity and core competitiveness of the enterprises.
RISTI, N.º E5, 07/2016 45

Secondly, based on the processes of the project application, project approval, financial
support and achievement transformation of the non-governmental economy in Suzhou,
the government should provide support and fund for the independently innovated
products. And the industrialization process for the products of the non-governmental
economy should be promoted. Meanwhile, relevant guides are expected to be formulated
to carry out the policy of purchasing the independently innovated products of the non-
governmental enterprises preferentially through the governments. By applying the
market demand as the guide, enterprises are encouraged to integrate with the universities
and scientific research institutions, so as to learn from each other’s advantages and carry
out associated research and development. The advantages of the emerging media such as
internet and twitter are supposed to be fully utilized, so as to create the high-tech sharing
platform that integrates information interaction, combination of industry-university-
research and achievement transformation. By doing so, the enterprise innovation can be
promoted by the three parts jointly.
3.4. Increasing the Efficiency of Energy Utilization and the Comprehensive

Utilization of Resources
Firstly, strict indexes and criterions require to be established in the processes of applying
and approving the projects. In this way, the initiation and development of the industries
with high energy consumption and pollution can be completely avoided, so that high
energy consumption and pollution are controlled from the sources. Meanwhile, strict
energy audit needs to be conducted on the heavy industry related projects that have
been implemented, so as to observe the energy consumption of the project constantly.
Meanwhile, to highlight energy conservation and transformation, the heavy industry
related projects are encouraged to utilize new energy of pollution free. Besides,
environmental protection departments have to carry out the environmental assessment
strictly, and release the assessment results in public. As to the projects or industries fail
to qualify the environmental assessment, the environmental protection departments
should supervise and urge them to rectify and reform within the deadline. If the projects
or industrials still cannot reach the standards after the rectification, they are ordered to
be closed. By doing so, the energy efficiency level of the non-governmental enterprises in
the industries with heavy energy consumption is further improved. This aims to make the
energy consumption per unit products reach the domestic advanced level, and encourage
enterprises to compete for the “Energy Star”.
Secondly, the enterprises should focus on promoting the source reduction, recycling,
remanufacturing, zero emission, and the development and application of industry
ecologicalization. Meanwhile, circular economy is encouraged to be developed in the
fields including industries, services, modern agriculture and vein industry to form a
batch of circular economy-based model enterprises. In addition, cleaner production
is supposed to be comprehensively implemented especially in the chemical, brewing,
papermaking, electroplating, textile, dyeing, building materials, steel industries and so
on. This aims to develop a batch of cleaner production-based model enterprises, and the
enterprises where resources can be comprehensively utilized.
46 RISTI, N.º E5, 07/2016

4. The Overall Support for the Development of the Non-governmental

Economy in Suzhou
4.1. Talent Environment
The introduction and cultivation of high-quality talents need to be accelerated.
Enterprises and governments have to actively take advantage of the activity platforms
such as “the week of the international elite in Suzhou”, “annual meeting concerning
the development of the merchants in Suzhou”, “win in Suzhou”, “plan for introducing
overseas talents” and “talent introduction plan on campus”. Meanwhile, excellent
management and technical personnels at home and abroad require to be introduced
actively. And relaxed environment and preferential policies are provided for them so
as to cultivate the excellent non-governmental entrepreneurs and establish the efficient
teams of non-governmental enterprises. Particularly, the leading personnels playing
an important part in the innovation and entrepreneurship of the emerging strategic
industries such as new energy, new medicine and biotechnology need to be introduced.
In this way, the advanced international experience and superb technical level possessed
by them can preferably serve the non-governmental economy in Suzhou. Meanwhile,
non-governmental enterprises have to widen their channels in employing, utilizing and
retaining talents. In a word, the innovation, transformation and upgrading of the non-
governmental enterprises can be promoted by creating a favorable talent environment.
4.2. Financial Environment

Multiple financing channels need to be explored. At present, financing difficulty is still
the maximum bottleneck restricting the rapid development of the mall and medium
non-governmental enterprises. According to the relevant surveys, most of the non-
governmental enterprises raise funds through borrowing from friends, internal fund-
raising of the enterprises, non-governmental financing load and interbank funding
between enterprises. These irregular financing channels present high risks. Therefore, it is
important to vigorously exploit the financial channels of various types, and comprehensively
give play to the functions of the credit financing, equity financing and debt financing,
as well as further widen the financing policies. All of these factors determine whether
or not non-governmental enterprises can realize the rapid development. When it comes
to the credit financing, commercial banks are encouraged to provide the corresponding
supports regarding the discount interest or risk compensation for the loan of mall and
medium enterprises. Meanwhile, financial institutions are encouraged to develop the
financial derivatives which are suitable for the characteristics of the non-governmental
enterprises, and increase the credit load placement in small enterprises. Meanwhile,
guarantee industries have to be standardized strictly, and supervised comprehensively
in terms of the industry permission, business operation, credit rating and industry
management. This can ensure that the guarantee scale is increasingly enlarged, and the
standards of business operation are strictly implemented, so as to promote the healthy
and ordered development of the guarantee industry.
RISTI, N.º E5, 07/2016 47

4.3. Government Environment

It is necessary to create a powerful government environment and accelerate the
construction of the public service platforms. Great quantity and high quality of non-
governmental enterprises are required by the “New Normal”. Accordingly, the
governments of various levels are required to actively promote the function transformation,
and correctly deal with the relationships between the government and the markets and
enterprises in particular. The purpose is to cultivate the entrepreneurship and promote
the transformation and upgrading of the non-governmental economy. As is well known,
the non-governmental economy in Suzhou has not been treated fairly in terms of the
tax preference in the past few years. Therefore, relevant government departments have
to give more positive supports to the capital financing, information consultation, talent
introduction and technical progress. Meanwhile, the governments need to establish
the joint conference system and hold meetings among all the member units in leading
groups of the non-governmental economy. In this way, the key and difficult problems
existing in the transformation and upgrading of the non-governmental economy can
be discussed and negotiated regularly. Meanwhile, the governments require to be
collaborated with the enterprises, chambers of commerce and associations to establish
the service platforms including information publishing, technical exchange, quality
evaluation and management consultation. Besides, based on the principles of resource
sharing, openness and compatibility, specialized services are expected to be provided for
the non-governmental economy, especially for the emerging strategic industries, so as to
satisfy the common demands of the non-governmental economy.
Acknowledgments
This work was supported in part by a grant from National Natural Science Foundation
of China (Number: 71002017). This work was supported in part by a grant from
Philosophical and Social Science Foundation of University in Jiangsu Province
(Number: 2014SJB548). This work was supported in part by a grant from “333” Projects
Foundation of Jiangsu Province. (Number: BRA2015417).
References
Chen, X., Chen, R. (2009). The Evaluation and Prediction of the Competitive Power
of Private Business in Changzhou, Suzhou, Wuxi and Nantong. Jiangsu Social
Sciences, 2, 219–226.
Esteban, J., Ray, D. (1994). On the Measurement of Polarization.Econometrica, 62,
819–851.
Freixo, J., & Rocha, Á. (2014). Arquitetura de Informação de Suporte à Gestão
da QualidadeemUnidadesHospitalares. RISTI-RevistaIbérica de Sistemas e
Jiang, X. (2009). Exploration on the Way of Realizing Scientific Development in Suzhou.
China Collective Economy, 9, 64–65.
48 RISTI, N.º E5, 07/2016

Lu, Y., Chen, X. (2013). Analysis on Operational Efficiency of Private Economy.

Enterprise Economy, 6, 97–100.
Pan, S. (2006). Debate on the Implication of Non-state Economy and Comment on It.
Economic Survey, 6, 85–87.
Qian, J. (2013). Research on the Fiscal and Taxation Policies to Promote the
Transformation of the Private Economy. Public Finance Research, 8, 59–63.
Zhang, L. (2011). Development Problems of Private Business in Suzhou under the
Transition economy. Modern Business, 30, 61–63.
RISTI, N.º E5, 07/2016 49

Effectiveness of College Counselors’ Work and

Ideological Education Based on Multimedia Platform
Pingqian Guo 1, Yafeng Zhang 2,*
* szxyzyf@163.com
1
School of Chemistry and Chemical Engineering, Suzhou University, Suzhou city, Anhui province,
234000, China
2
Pages: 50–60
Abstract: With the development of electronic information, ideological and political

education presents a new feature. At the same time, the influence of the network
is gradually infiltrated into political education, which provides opportunities and
challenges for the ideological and political education. In this paper, the author test
the effectiveness of counselors’ work based on multimedia platform data, results
show that 85% students think the counselor has great influence to their mental
health and future career development, and 83.3% counselors feel satisfied with
their work. In order to improve the effectiveness of the counselors, college should
establish a clear, scientific management system and professional development
mechanism. Also, college should strength the responsibility of counselors, and
enhances the work loyalty.
Keywords: Network education, college counselors, multimedia platform,
ideological education
1. Introduction
Electronic media will bring revolution to our society, with the arrival of the era of
electronic information, ideological and political education presents a new feature.
At the same time, the influence of the network is gradually infiltrated into the
process of Ideological and political education, which provides opportunities and
challenges for the ideological and political education (Mastroleo, 2010; Carrell,
2014). Correctly grasp the characteristics of Ideological and political education in
the network environment, seize the opportunity, meet the challenge, innovate the
ideological and political education, and become the era subject of the ideological and
political education in the network environment. College counselors and teachers are
the backbone of College Students’ Ideological and political education, and play an
important role in the ideological and political education of students. To strengthen
the team building is an organizational guarantee for the ideological and political
education of college students (Huang, 2013; Sá, 2012; Alavi,2012). Questionnaire
survey shows that our instructors in Colleges and universities in the age structure, the
school record structure, subject structure, position / Title structure showing a certain
50 RISTI, N.º E5, 07/2016

characteristics; types of counselors is diversified and to the construction of professional

counselors. To improve the counselor and teacher troop overall quality must provide
corresponding policy, security conditions, strengthen the construction of the team.
Network education is to actively integrate the advantages of the network resources for the
ideological and political education, the establishment of professional web sites to provide
a solid technical support, and create a positive network culture atmosphere (Khansa,
2015; Catts,2015). To the advantage of relying on the campus culture construction,
the informatization, digitalization and network construction of embedded in the
ideological and political education system, through the network to achieve full coverage,
full integration, full penetration of the correct ideas of communication theory. At the
same time, to create a is familiar with the work of Ideological and political education,
and proficient in the dissemination of knowledge, is good at network technology of the
composite type of Ideological and political education teachers, is particularly important
to grasp the initiative in the construction of network environment (Anand, 2010; Michel,
2012; Sabella, 2010). High quality ideological and political teachers should have a high
level of political theory, can skillfully apply the communication skills to the ideological
and political education activities, and skillfully use the network to guide the correct
direction of public opinion.
2. Multimedia learning platform

2.1. E-learning platform
The ideological and Political Education under the network environment created by
the communication and interaction with extensive model of openness and contact
information, make the educated can according to the needs of active search, take the
initiative to seek, education of “push” become educated “pull”. The media resources of
Ideological and political education in the network environment are open. Multimedia
classroom as shown in figure 1 and content of E-learning platform as shown in figure 2.
Figure 1 – Multimedia classroom

The network environment brings both opportunities and challenges to ideological and
political education. First, the network has expanded the position of Ideological
and political education. The special medium of the network has the ability of
mass information storage, and its nonlinear spread has the characteristics of wide
coverage, so it can enrich the content of the ideological and political education of the
network. In addition, traditional and single way of Ideological and political education
in classroom and book as the basis, relatively fixed location, narrow coverage of
RISTI, N.º E5, 07/2016 51

Effectiveness of College Counselors’ Work and Ideological Education Based on Multimedia Platform
Figure 2 – Content of E-learning platform
information, the influence is small, the network to make ideological and political
education get rid of the constraints of time, to improve the students’ participation
and initiative; illustrated, intuitive image information, the effect of the ideological
and political education is also greatly enhanced. Moreover, under the environment
of network ideological and political education on the subject and object of duality,
the educators in the interactive network, can quickly grasp the latest and most
comprehensive education information feedback, greatly enhancing the effectiveness
of Ideological and political education work, to further strengthen the educational
function of Ideological and political education.
2.2. Counselors work under network environment

Network information technology has a revolutionary impact on the development of
Ideological and political education, must be highly valued. The network breaks the
boundaries between the real world and the virtual world, and creates a good atmosphere
for the ideological and political education to create an equal, open and democratic
dialogue. In the traditional sense, the identity of the subject and object of education in the
network environment shows more dual nature and anonymity, which makes the equal
status of education both have been further strengthened, narrowing the psychological
distance of their communication. Network between the educators and the educated,
set up a communication bridge, it can be to exclude the interference factors under the
condition of the educated are more willing to adopt the equal dialogue and education,
giving vent to their true feelings. Educators can also interact with the educated in the
process of equality, Multimedia learning principles as shown in figure 3.
52 RISTI, N.º E5, 07/2016

Figure 3 – Multimedia learning principles
3. Research design and variables

3.1. Network questionnaire
Strengthening the construction of Ideological and political education work team is
an organizational guarantee to strengthen and improve the ideological and political
education of College students. Now, the main body of the ideological and political
education of college students is the school party and government cadres and the
Communist Youth League cadres, ideological and political theory course and the study of
social science teachers, counselors and teachers. Among them, the instructor, the teacher
in charge is the moral education in Colleges and universities to carry out the ideological
and political education of the backbone of the force. Therefore, the understanding of the
current college counselors, the basic situation of the class teacher team, to study their
characteristics, to effectively strengthen the construction of college counselors, class
teachers has a very important significance.
Figure 4 – Network questionnaire

Through the network questionnaire, we choose 600 counselors as research samples.
Among them, 15 undergraduate colleges and universities, the instructors population
RISTI, N.º E5, 07/2016 53

accounted for 40%; 13 in higher vocational colleges, instructor’s population accounted

for 33.9%; private colleges, instructor’s population accounted for 26.1%.
3.2. Current situation of College Counselors

College counselors, the age structure of the class teacher team is the basic state of the
team, it reflects the survival and development of the team, see table 1. These cases
illustrate that one aspect the present stage in China’s colleges and universities counselors,
head teachers younger degree is high, at the same time that they are a with vigor and
vitality, energetic, young and energetic team; on the other hand, existing university
counselor team from thing students ideological and political education work experience,
experience and they engaged in university teachers or related work experience is closely
related to. Questionnaire survey showed that the case, just graduated from college in
counselors, the work of a teacher in charge of a class, about 20%, they are engaged in the
ideological and political education work experience with them in college work experience
almost simultaneously can say they are in “learning by doing”. In a sense, the rich long-
term engaged in the ideological and political education of students work experience and
experience of counselors, teacher in charge, with the growth of the age and the number
of less and less trend.
instructor Age number percentage
youth 22-25 years old 108 18.5%

middle age 26-30 years old 257 43.6%
old age 31-35 years old 132 22.6%

youth 36-40 years old 43 7.3%
41-55 years old 30 5.0%
middle age
46-50 years old 18 3.0%
Table 1 – Age distribution of College Counselors

Counselors, teachers should have higher education levels than other teachers in general,
which is adequate for the work of college students to work, the objective needs of
ideological education work. At present, the academic structure of college counselors and
teachers in charge of the class is mostly college degree. Among them, with a bachelor’s
degree accounted for 66% of the survey, with a graduate degree as 31.8%; with the
proportion of doctoral graduate degree is only 2.2%. Thus, although the current college
counselors, the master’s degree in line with the basic standards, but there is a gap
between the requirements of college instructors, see table 2.
education number percentage

undergraduate 396 66.0
graduate student 191 31.8
doctor 13 2.2
Table 2 – Educational structure of College Counselors
54 RISTI, N.º E5, 07/2016

According to the results of questionnaire survey, the subject structure of college counselors
and head teachers in Colleges and universities is in line with the situation of college type.
Overall, most of the professional training of counselors, class teacher on the ideological
and political education to do a good job in the preparation of knowledge related to lack of
knowledge. Therefore, it is urgent and important to strengthen the training work of the
counselors and teachers in charge of the class. From the results of our survey, the current
counselor, teacher team of subject structure has the following characteristics: has the
philosophy and social science disciplines and subjects accounted for 9.6% of investigation
number, with education, psychology and sociology majors accounted for 3.8%, the subject
designed industry and related to counselors professional requirements closer to, but
both add up to only 13.4%, relatively few. With humanities disciplines (16.7%), with the
economics and management disciplines of 12.9%, with science and engineering majors
accounted for 22.5%, according to counselors in Ideological and political education of
college students need professional knowledge structure, we need to give them related
disciplines of training and learning. Details of the survey are shown in table 3.
education number percentage

Philosophy 55 9.6%
Education and psychology 23 3.8%
language 101 16.7%
Economics & management 224 37.9%
Polytechnic 135 22.5%
other 61 10.4
Table 3 – Subject structure of the instructor
4. Effectiveness of counselors’ work

4.1. Job evaluation
The primary object of the work of the counselors is the students; the basic goal is the
comprehensive development of students and talent. So, evaluation of the effectiveness of
the work of counselors, the most have the right to speak the students should be. Survey
shows that students think that counselors are college / University contact most, for their
help the greatest that counselor’s work of self-growth of physical and mental health
and future career development has great guidance and guidance. Among them, 85%
students think that the influence of personal counselor, counselors work is very effective,
promotes the growth and ability of his thoughts, help yourself to shape the good outlook
on life; 11.5% students think of counselors in general, just now life trivia and habits form;
only 3.5% students think the effectiveness of counselor work basic is not reflected. By the
difference analysis it is found that student cadre of counselor effectiveness evaluation
of higher than non-student cadre, think counselor is his mentor and close friend of the
student cadres accounted for 74.5% those counselors is to guide the work of student
cadres accounted for 15%. In addition, a student of counselor effectiveness evaluation is
generally higher than that of students in other grades; they think counselor is to adapt
herself to the primary guide for college life.
RISTI, N.º E5, 07/2016 55

Student grade
Evaluation number percentage
Grade 1 Grade 2 Grade 3 Grade 4
Very effective 510 85% 245 160 103 92
commonly 77 11.5% 32 24 15 6
No effect 13 3.5% 2 3 4 3
Table 4 – Evaluation of students’ work

Counselors work is effective, in addition to student evaluation, the Counselor’s
self-awareness and self-evaluation is also very important. From the Counselor’s self-
evaluation data, most of them are “more satisfied” and “general”, said “very satisfied”
only 2.1%, said “not satisfied” and “very dissatisfied” only 7.5%. From the subjective
evaluation, the majority of counselors believe that they have the basic ability and
quality, can be competent for the work of counselors. Among them, 26% counselors said
that “fully qualified” this work, 64.8% counselors said “basic competence”, 8.55% the
counselors said “barely competent”, and 0.65% of the counselors said “not competent”.
Influenced by the social environment and the development ideas and ideas of higher
education, the college counselors are still at the edge of the University, and the
recognition of the effectiveness of the work is not high. First of all, the importance of
College Leadership on the work of counselors is not high. Think that the work of college
counselors is very important in 20.4%, that the work of college counselors is more
important than 36.5%, that the work of College Counselors in general and the following
43.2%. In Colleges and universities, the importance of leadership to teaching is generally
higher than the degree of attention to the work of counselors. Although most colleges and
universities believe that the work of counselors is indispensable, but when the school has
a major policy choices and development choices, the work of the counselors are always
pushed or ignored. Secondly, the other administrative departments of the education and
ideological understanding of the work of counselors is not enough, it does not recognize.
Most departments believe that the instructor is only responsible for the implementation
of the Department’s work arrangements, the implementation of good, good is effective,
and does not require the work of innovation. Again, full-time teachers, work of counselors
there discrimination that instructors lack of professional theory and ability to conduct
scientific research and teaching research, as long as the discipline of students do a good
job is effective. In addition, although the state policies and regulations of teachers and
administrators have dual identity, but in many colleges and universities, counselors are
excluded from the teachers, a lot of relevant treatment and policy cannot enjoy.
evaluation number percentage

Fully qualified 156 26%
Basic competence 384 64.8%
Barely qualified 50 8.55%
Not competent 4 0.65
Table 5 – Counselors’ self-evaluation
56 RISTI, N.º E5, 07/2016

4.2. Factors affecting the effectiveness of College Counselors’ work

The effectiveness of the work of college counselors plays an important guiding role
in promoting the growth of college students, maintaining the stable development of
colleges and universities. But due to the influence of subjective and objective factors, the
effectiveness of counselor work haven’t been fully implemented, did not fully play its role
as the core of Ideological and political education, did not obtain the development level
and status and teaching the same. Through investigation and analysis, the reasons that
influence the effectiveness of the work of college counselors mainly include: subjective
factors of counselors’ professional value orientation, education, personal qualities, etc.;
the execution of national policies, counselor team management and counselor career
development system, objective factors. The objective factor is the key to determine the
effectiveness of college counselors’ work at present. Subjective reasons as the lack of life
value orientation of higher vocational attitude, value orientation is refers to the subject of
value based on his view in the face and deal with all kinds of contradictions, conflicts and
relations to basic value of the position, attitude and value and the basic value orientation.
The overall value of student work concept is to “educating people” as the ultimate goal.
The survey data show that the positive effectiveness of counselors occupation attitude and
related work. The subjective is more willing to engage in the occupation of the effectiveness
of counselors’ work satisfaction is higher. Among them, “very happy” to choose the
occupation counselors, 83.3% of people said it was “very satisfied or satisfied with my job”.
“Not willing” or “very happy” to choose this occupation counselor, only 11.1% of people
said it was “satisfactory” in their work, feel very satisfied with the people there. Thus, the
subjective value orientation of counselors is an important factor affecting the effectiveness
of counselors’ work. From this investigation can see that most of the counselors targeting
education is relatively low (see Table 6), completely ignore the education uniqueness
of counselor work, did not distinguish between Counselors and other teachers in the
educational function of the difference. The most fundamental reason lies in the fact that the
target of the counselor education deviates from the highest level of educational philosophy.
sort goals number percentage

1 Let students have mental health 233 43.1%
2 Let students become law-abiding people 117 21.7%
3 Let students become knowledgeable people 109 20.2%
4 Let students have good moral quality 68 12.6%
5 Let students to set up the lofty ideal of communism 12 2.2%
Table 6 – Educational goal orientation of Counselors

Counselor’s personal accomplishment structure is not perfect, a person’s quality
performance in many aspects, mainly through the individual in life, work and other
aspects of behavior and way of doing things. The personal qualities, which are the most
closely related to the effectiveness of the instructors, include ideological and political
quality, moral quality, psychological quality, knowledge theoretical quality, ability
quality and physical quality. Stress quality has been put in the most important position,
and the counselors’ Ideological and political quality, moral quality, psychological quality
and other personal qualities have not received more attention, no better performance
RISTI, N.º E5, 07/2016 57

of the actual value. Counselors individual quality structure is not perfect, resulting
in counselor in philosophy, education methods and personality exist certain defects,
concept of value for students of education and training of the impact and effectiveness of
counselor work difficult to achieve the essential purpose of education.
The objective reason is the key to the effectiveness of College Counselors’ work at
present. It can be said that the constraints of the objective factors lead to the generation
of subjective factors. Among them, the most important of the three objective reasons is
staffing and future career development of counselor of the division of responsibilities and
counselors. Staffing imbalance leads to low work efficiency, college counselors personnel
equipped with low proportion, counselors with number of students overall side. As can be
seen, the number of students with 92% of the counselors does not meet the requirements,
the number of people outside the Department of education. And the number of students
with too much, will inevitably affect the effectiveness of education and management of
counselors, and reduce the effectiveness of the work of counselors. Survey shows that
the number of students and counselors work effectiveness was negatively correlated.
Among them, with 200 students under the counselor to work effectively said “very
satisfied” or “more satisfied” accounted for 76%, said “dissatisfied” or “very dissatisfied”
0; with more than 500 students counselors to work effectively expressed “very satisfied”
or “relatively satisfied” with the proportion as 23.2%, expressed “dissatisfied” or “very
dissatisfied” proportion was 17.7%. As can be seen, with the increase in the number
of students, the effectiveness of counselors work is gradually reduced, the number of
students is an important factor affecting the effectiveness of the work of counselors.
The core function of the instructor is to cultivate students’ Ideological Education
and ability, to guide students to establish a correct world outlook, outlook on life and
values. But the survey found that the time distribution of the daily work of counselors
deviated from the core functions, resulting in a functional alienation. In this survey,
the distribution of weekly working hours of the counselors were classified and sorted,
as shown in table 7.

1 Managing student affairs 318 44.3%
2 Administrative affairs 183 17.5%
3 Students ability training 136 14.6%
4 Ideological Education 105 13.3%
5 Psychological counseling 132 8.1%
6 Learning Guide 65 5.2%
Table 7 – Daily work order of Instructors

Instructors engaged in too much work, have no time to take into account the student’s
ideological and political education, psychological education, career planning guidance, such
as the core work. This not only resulted in the alienation of the work of the counselors, but
also affected the effectiveness of the work of counselors. Survey shows that in the student
management affairs and administrative work of counselors, on the effectiveness of the
said satisfactory ratio were 48.6% and 26.2%, expressed not satisfied with the proportion
58 RISTI, N.º E5, 07/2016

of 9.9% and 39.9% respectively; while the time spent in the students of Ideological and
political education and ability training instructor, expressed on the effectiveness of the work
satisfaction rates were 61.2% and 62.3%, The proportions of the expressed dissatisfaction
were 4.1% and 2.1% respectively. It can be explained that the reasonable distribution of
the daily working hours and the correct exercise of their work responsibilities are the key
to the effectiveness of the counselors’ work. From the order of table 8, we can see that the
most hope for the reform of the mechanism is also the main objective difficulties faced by
college counselor’s work. Therefore, to fundamentally improve the effectiveness of the work
of counselors, the reform of the management system for counselors, to create a clear and
smooth development of the mechanism is the key.

1 Scientific division 355 21.9%
2 Improve work treatment 322 19.9%
3 career development mechanism 278 17.2%
4 Increase training opportunities 259 16%
5 Clarify the job responsibilities 153 9.4%
6 Enhance the value identity 149 9.2%
7 Strengthen examination management 63 3.9%
8 psychological counseling courses 40 2.5%
Table 8 – Content of school reform
5. Conclusions
In order to improve the effectiveness of counselors’ works, we need to improve the
quality of the subjective and the objective of the management system. On the one hand,
instructor to improve the effectiveness of self-work, set counselors undertaking the value
pursuit of the ideal, the pursuit of external urge to internalize the occupational interest
and value; on the other hand, a clear, scientific management system and professional
development mechanism, do “management guide, management development. Effective
ideological and political educators and educational objects should have the effectiveness
of including potential energy, kinetic energy dimension two dimensions. Embodied
in dimensional potential energy efficiency of the ideological and political education of
Ideological and political education in the strong subject consciousness, cause of the
ideological and political education work and sense of responsibility, kinetic energy
efficiency as a concrete manifestation of the ideological and political education of political
accomplishment, personality, theoretical preparation, teaches education practice ability.
Therefore, to enhance the effectiveness of counselors work, the first is to start from the
two dimensions of the potential energy and the kinetic energy dimension.
First of all, school should establish the career of the career concept. Through publicity,
strengthen the sense of responsibility and sense of mission counselors, counselors to
enhance the cause of loyalty. Secondly, enhance the professional interests and the sense
of happiness, and create an atmosphere of recognition and respect for the instructor
in the school. In the policy formulation should respect the opinions of the Counselor
RISTI, N.º E5, 07/2016 59

Group, in the work to affirm the value of the work of counselors and results, so that the
instructor group in the school has a sense of status and honor. To implement flexible
management, strengthen the humanistic care of the instructor. We should actively expand
the development platform for counselors, to provide a chance to show self talent, realize
the value of self, so that the counselor to obtain a lasting career happiness. Secondly,
in order to improve the teaching practice ability of counselors to create ways. Through
the activities of external communication, business learning and training, counselors
and other ways to strengthen the training of counselors and other ways to expand the
exchange of counselors. Again, relieve the pressure of the counselor’s career. Through
the organization investigation, establishment of counselor club, carry out the campus
cultural activities, organization experience the outdoor type development activities and
counselor forum and Sharon and other ways to carry out, for counselors provide diverse
relieve pressure and emotion regulation of the platform.
References
Alavi, M. (2012). The Challenges of High School Counselors in Work Place Original.
Procedia - Social and Behavioral Sciences, 46, 4786–4792.
Anand, K. (2010). Context-general and Context-specific Determinants of Online
Satisfaction and Loyalty for Commerce and Content Sites. Journal of Interactive
Marketing, 24, 222–238.
Carrell, S., Hoekstra, M. (2014). Are school counselors an effective education input?
Economics Letters, 125, 66–69.
Catts, Z. (2015). Certified Genetic Counselors: A Crucial Clinical Resource in the
Management of Patients with Suspected Hereditary Cancer Syndromes. Surgical
Oncology Clinics of North America, 24, 653–666.
Huang, Z., Benyoucef, M. (2013). From e-commerce to social commerce: A close look
at design features. Electronic Commerce Research and Applications, 12, 246–259.
Khansa, R. (2015). Teachers’ Perceptions toward School Counselors in Selected Private
Schools in Lebanon. Procedia - Social and Behavioral Sciences, 185, 381–387.
Mastroleo, N. R., Turrisi, R. (2010). Examination of posttraining supervision of peer
counselors in a motivational enhancement intervention to reduce drinking in a
sample of heavy-drinking college students. Journal of Substance Abuse Treatment,
39, 289–297.
Michel, L., Mohammad, R. (2012). The effects of social media based brand communities
on brand community markers, value creation practices, brand trust and brand
loyalty. Computers in Human Behavior, 28, 1755–1767.
Sá, F., & Rocha, Á. (2012). Definição da arquitetura de informação em organismo da
administração pública local. RISTI-Revista Ibérica de Sistemas e Tecnologias de
Informação, (10), 51–64.
Sabella, R. (2010). School counselors perceived importance of counseling technology
competencies, Computers in Human Behavior, 26, 609–617.
60 RISTI, N.º E5, 07/2016

The Influence of Strategic HRM on Enterprise

Performance based on Online Survey and Data Mining
Shuxia Wang 1,2
szxywangshuxia@163.com
1
School of Management Engineering, Suzhou University, Suzhou city, Anhui province,234000, China
2
School of Management, University of Science and Technology of China, Hefei city, Anhui province,
230026, China
Pages: 61–70
Abstract: With the development of information technology, data mining

technology is widely used in business survey and analysis. In this paper, the author
analyzes the influence of strategic HRM on enterprise performance. By using
structural equation method, the result shows that strategic HRM has positive
influence on enterprise performance, the path coefficient is 0. 546. At the same
time, staff training and the implementation of the performance salary is played the
adjustment effect. In conclusion, human resource is the carrier of innovation; the
strategic HRM can make the enterprise obtain the humans that meet the strategic
target and then produce the innovation performance.
Keywords: Data mining, strategic HRM, structural equation, employee training
1. Introduction
With the development of human resource management practice, management scholars
and practitioners pay more and more attention to the strategic value of human resources
(Jenster,1999; Zafar, 2013). Due to the acquisition, development and retention of human
resources is embedded in the unique history and culture of the enterprise, so it is the source
of enterprise’s unique competitive advantage. In order to realize the strategic value of
human resources, the management and operation of human resources must have a strategic
orientation, the enterprise needs to carry on the strategic human resources management
(David, 1997; Stone, 2015). Martell and Carroll in the 89 Fortune 500 companies in the
process of research found that 40% to 69% subsidiaries of the human resources department
to a certain extent, involved in the process of enterprise strategic management. Strategic
human resource management has always been an important issue in the research of human
resource management. The research shows that strategic human resource management
can improve the performance of the enterprise by cultivating the employee’s behavior in
line with the requirements of the organization strategy, promoting the employee’s behavior
consistent with the organization’s goal (Gree, 1999; Pereira, 2015).
Strategic human resource management will change the characteristics of enterprise
employees and shape the enterprise innovation culture, in order to improve the
RISTI, N.º E5, 07/2016 61

The Influence of Strategic HRM on Enterprise Performance based on Online Survey and Data Mining
enterprise’s knowledge creativity, and thus improve the enterprise performance.

Strategic human resources management will be effective in the management of human
resource management and enterprise strategy to match the formation and development
of enterprise culture within the enterprise, and thus enhance the performance of
enterprises (James, 2013; Nagendra, 2014). Research shows that human resource
is the carrier of innovation, the strategic human resource management can make
the enterprise obtain the human resources that meet the strategic objectives of the
enterprise to carry on the enterprise innovation, and then produce the innovation
performance. Strategic human resource management will be in the enterprise interior
create innovative atmosphere, motivate employees to assume more risk and implement
positive “active” market, such as research and development strategy, promote the design
and implementation of internal corporate entrepreneurship strategy, build long-term
competitive advantage of enterprises (Hussain,2007). Based on the study of system
theory that different institutional environment, the influence degree of the enterprise
strategic human resource management of enterprise performance different, unfair
competition environment will hinder the strategic human resource management on firm
performance have positive effects and the support environment significantly strengthen
the strategic human resource management and enterprise performance between role
in promoting. On the basis of existing research, this study through career development
and employee training plan, performance salary, temporary staff management, and
other typical employee participation in management practice to examine the influence
of strategic human resource management on enterprise performance mechanism.
2. Data mining and prediction

2.1. Data mining
In the complex target, the analysis of massive data sets, there is no ready-made and can
meet the conditions of the general theory of the method of calculation. However, the neural
network has a great advantage in the ability of the high bearing capacity of the noise data
and the ability to classify the training data. Therefore, it is feasible and necessary to design
a data mining method based on neural network, and to use it in real world problems.
Artificial neural network can be used in data mining classification, clustering, feature
mining, prediction and pattern recognition and so on. Therefore, artificial neural network
plays an important role in data mining, the neural network hierarchy as shown in figure 1.
Figure 1 – The neural network hierarchy
62 RISTI, N.º E5, 07/2016

Neural Network Artificial has a strong computing power. First of all, there is a huge
parallel distributed structure; secondly, it has the ability to learn and the resulting
induction. Data mining refers to the process from a large number of incomplete, noisy,
fuzzy, random data, extracted implicit in which people do not know in advance, but also
potentially useful information and knowledge of the process. It is an effective means
of knowledge discovery. Discovered knowledge not only can be used for information
management, query optimization, decision support, process control, etc., but also can
be used for the maintenance of the data itself.
Figure 2 – data mining process
2.2. Data prediction

During the process of data mining, data mining is the core content of the whole process:
•• Classification and prediction: tutor supervision learning to find a group
to be able to describe the data collection of the typical characteristics of the
model, in order to be able to classification and recognition of unknown data
belongs to a category is unknown cases mapping to a discrete categories. The
main representation methods include: classification rules, decision trees,
mathematical formulas and neural networks.
•• Clustering analysis: unsupervised learning methods, according to the
different aggregation between internal data object of maximizing the similarity,
and the aggregation minimizing the similarity between objects “as the basic
clustering analysis principle, the clustering analysis of data object is divided into
several groups. Each cluster analysis to obtain the group can be considered for a
similar don’t belong to the set of data objects, further from the same data set, and
can obtain the corresponding classification prediction model (rules) through the
classification study. In addition, a hierarchical model of the initial data set can
be obtained by the repeated clustering analysis of the obtained clustering group.
•• Outlier analysis: the analysis of isolated point data, in which the data is
usually excluded from the data mining analysis of data mining analysis of
the data.
RISTI, N.º E5, 07/2016 63

Figure 3 – Data prediction based on GMDH Shell
3. Research assumption
Huselid put forward some of the human resource practices known as “best practices”
are often superior to other human resource practices, thus promoting the improvement
of organizational performance. Employees are actively involved in the organization
of the business to expand the autonomy of employees, by stimulating their internal
needs to increase the input of employees. So let employees to participate in enterprise
management practice, improve the activities of the enterprise strategic participation
degree and its ability, so as to make the enterprise by improving the effectiveness of
strategic management of strategic human resources management are the effective
supplement. When the enterprises to carry out the strategy of the new human resources
by allowing employees to participate in the decision-making process, and let them know
the organization’s current business information to improve the staff to the acceptance
of the new HR strategy and investment. Trade unions participate in the sharing of
detailed information about strategic decisions made by employers and employees to
work together to improve performance. Career development and training opportunities
can help employees acquire the ability to implement strategies. Performance based on
personal interests and the company aims to motivate employees to achieve the strategic
objectives of human resource management. There are many temporary jobs (for example,
contract workers and temporary workers) in the enterprise, the cooperative relationship
between temporary and regular employees is the key to the effective implementation
of strategies. Therefore, we think that a variety of typical let employee opportunity
to actively participate in management practice, such as union participation, career
development and staff training plan, pay for performance, temporary staff management
etc., strategic human resources management more efficiency. But given to Chinese
enterprises especially private economies, trade unions are often non-existent, even in
large state-owned unions are often just collective activities of the organizers or material
welfare payment, trade unions to participate in the formulation of the strategy of the
64 RISTI, N.º E5, 07/2016

possibility is very small, so we will mainly discuss the career development and staff
training plan, pay for performance, temporary staff management three typical employee
participation in management practice activities can be of assistance in strategic human
resource management play effectiveness and impact on enterprise performance.
3.1. Career development and employee training program

Enterprise’s unique human capital is the fundamental source of competitive advantage.
In order to cope with the changing environment and gain sustainable competitive
advantage, enterprises must develop flexible human capital to quickly adapt to the
possible occurrence of enterprise strategic change. Career development plans to expand
the capacity of employees, to provide support for career development. Employee training
programs give employees the opportunity to acquire skills and knowledge to improve job
performance. Through career development and employee training programs, employees
are exposed to a wide range of knowledge and perspectives, so that they are no longer
resistant to change, and more willing to accept new knowledge and creative approaches.
In this regard, the rich career development and staff training extended the can be
used for the implementation of the strategy of human resources, and help enterprises
to flexibly adjust the human resource to meet various strategic demand. Therefore,
career development and employee training programs can expand the positive impact of
strategic human resource management on firm performance.
Hypothesis 1: employee career development and training programs will enhance the
positive impact of strategic HRM on firm performance.
3.2. Career development and employee training program

Enterprise’s unique human capital is the fundamental source of competitive advantage.
In order to cope with the changing environment and gain sustainable competitive
advantage, enterprises must develop flexible human capital to quickly adapt to the
possible occurrence of enterprise strategic change. Career development plans to expand
the capacity of employees, to provide support for career development. Employee training
programs give employees the opportunity to acquire skills and knowledge to improve job
performance. Through career development and employee training programs, employees
are exposed to a wide range of knowledge and perspectives, so that they are no longer
resistant to change, and more willing to accept new knowledge and creative approaches.
In this regard, the rich career development and staff training extended the can be
used for the implementation of the strategy of human resources, and help enterprises
to flexibly adjust the human resource to meet various strategic demand. Therefore,
career development and employee training programs can expand the positive impact of
strategic human resource management on firm performance.
Hypothesis 2: employee career development and training programs will enhance the
positive impact of strategic HRM on firm performance.
3.3. Temporary staff management

In recent years, some studies have pointed out that temporary employees can also
become strategic assets of enterprises. It is considered that the enterprise should
RISTI, N.º E5, 07/2016 65

adopt different modes and strategic use of human resources in different modes of
employment. They believe that there are two models for the management of temporary
employees: temporary employment contracts and alliances. In temporary employment
contract mode, if the enterprise human resource and is not very important or unique,
enterprises can consider hiring highly skilled contract workers, because hire their
costs may be relatively low, also the on-the-job employees formed competitive threat,
motivate them to work hard and improve their work performance. In the alliance mode,
enterprises and temporary employees to establish a partnership, so that their expertise,
knowledge and specific results for the enterprise. Anyway, temporary workers can
provide enterprises with professional skills and knowledge, some possible formal
internal staff do not have, so to the enterprise has very high value, has formed the
effective supplement of enterprise knowledge and capacity of the system. Therefore, it
can help enterprises to gain competitive advantage, improve enterprise performance,
and the cooperation of temporary employees, in turn, and support the implementation
of human resources strategy.
Hypothesis 3: temporary staff management system will enhance the positive impact
of strategic HRM of the enterprise performance.
4.1. Study sample characteristics
The sample of this study comes from 1245 enterprises in the eastern, central and
western cities of the city, and a total of 1245 questionnaires were issued to 100
enterprises. The questionnaire was completed by the person who is familiar with the
development situation of the enterprise and the development of the industry. A total of
1027 questionnaires were recovered. The recovery rate was 82.49%, and the effective
questionnaires were 874, and the effective rate was 85.10%. The measurement of the
variable is a part of the five stage of the use of the Li G. In order to ensure the validity
and reliability of the measurement, the study was carried out on the basis of a small
sample survey. As table 1 shows, strategic human resource management, including six
items and the scale Cronbach’α reliability coefficient is 0.89; measurement of temporary
staff management includes four items. The scale Cronbach’α reliability coefficient is
0.87, career development and staff training is the average number of enterprises with
formal employee participation in the career development and staff training project;
performance pay schemes is formal employees of the enterprise performance wages
accounted for the average proportion of the total wages; enterprise performance is
mainly used to measure business profits.
4.2. Research results

Firstly, the path coefficient of each variable is obtained by the method of structural
equation analysis. As shown in Table 2, strategic human resource management, career
development and staff training programs, performance pay, temporary employee
management have significant positive impact on enterprise performance. The hierarchical
regression analysis method was used to verify the hypotheses. Career development
and staff training plan, pay for performance, temporary staff management to strategic
66 RISTI, N.º E5, 07/2016

variable Survey item Reliability

coefficient
Human resource management closely with enterprise strategy
HRM has a great influence on the development of enterprise
Strategic strategy.
human
Human resource managers will greatly affect the decision making 0.89
resource
of enterprise CEO
management
HRM supports the realization of the strategic objectives
Has taken effective measures to ensure the coordination of HRM
Establish a complete management system for temporary
employees
Temporary Will establish a detailed management file for temporary staff 0.87
employees
Arrange appropriate training for temporary staff.
Performance evaluation of temporary employees
Table 1 – Variable measurement and reliability of measuring
human resource management and enterprise performance influence the effectiveness of

regulation by using the proposed by Baron regulated utility model validation procedures
to verify. Because of the difference between the different variables, the numerical
interval is not a, so we need to carry on the standard treatment of all variables before the
adjustment effect analysis is carried out. The min-max Standardization for moderating
effects of judgment is divided into the following two steps: first, the direct with existing
variables the dependent variables to make the regression analysis (model M1); then,
through the predefined variables and moderating variable is calculated by multiplying
a factor interaction, which is added to the independent variables for regression analysis
(model M 2). If the R2 value of the model M2 is greater than the R2 value of the model
M1, and the independent variables and the interactive factor regression coefficients are
significant, then the regulation effect exists; if not, it does not exist at the same time.
Relationship Path coefficient Significant degree

Strategic human resource management→ 0. 546 significant
enterprise performance
Career development and employee training → 0. 417 significant
business performance
performance-related pay→ 0. 412 significant
firm performance
Temporary employee management→ 0. 389 significant
enterprise performance
Table 2 – Influence coefficient between variables

Career development and staff training effectiveness of regulation analysis results are
shown in Table 3, increase the amount of R2 is 0. 32. At the same time, independent
variables and interaction factors were significant in the 0.05 confidence level. Therefore,
career development and staff training have positive influence on strategic human
RISTI, N.º E5, 07/2016 67

resources management and firm performance relationship exist utility regulation,

hypothesis 1 was verified.
regression model
independent
variables M1 Regression coefficient M2 Regression coefficient
and T value and T value
① Strategic HRM 0. 554(9. 28) ** 0. 332(6. 56)*
②career development
0. 467(8. 76) ** 0. 196(5. 76) *
and employee training
①×② 0. 132(5. 54)
R2 0. 23 0. 55
addition of R2 0. 32
Table 3 – Career development and employee training regulation effectiveness
The implementation of performance pay program regulated utility analysis results are
shown in Table 4, increase the amount of R2 is 0. 2. At the same time, independent
variables and interaction factors respectively significant at 0.01 and 0.05 confidence
levels. Therefore, performance pay program implementation of strategic human
resources management and firm performance positively related to regulate the utility,
hypothesis 2 was verified.
regression model
independent
①Strategic HRM 0. 536(9. 58) ** 0. 326(8. 64)**
②Merit pay 0. 401(4. 64)* 0. 208(5. 11)*
①×② 0. 129(4. 99) *
R2 0. 32 0. 52
Table 4 – Merit pay adjustment utility validation

Temporary staff management system regulated utility analysis results are shown in
Table 5, increase the amount of R2 is 0. 23. At the same time, independent variables
and interaction factors respectively at 0.01 and 0.05 confidence level significantly.
Therefore, temporary staff management system of positive influence relationship of
strategic human resource management and enterprise performance has a moderating
effect, hypothesis 3 was verified. Due to the assumption that 1, supposing 2 and
3 have been verified. Therefore, career development and staff training plan, pay for
performance, temporary staff management’s positive influence on strategic human
resource management and corporate performance relationship exist utility regulation.
At the same time, we can see the career development and staff training plan, pay for
68 RISTI, N.º E5, 07/2016

performance, temporary staff management of the effectiveness of regulation size, the

career development and training programs for staff, temporary staff management time,
pay for performance plan minimum.
regression model
independent
①Strategic HRM 0. 579(8. 98) ** 0. 309(8. 85)**
②Temporary staff
0. 452(5. 32)* 0. 186(3. 33)
management
①×② 0. 119(5. 62) *
R2 0. 26 0. 49
Table 5 – Temporary staff management
5. Conclusion
Human resource is the core strategic resource of the enterprise, and the enterprise human
resource management activities should be integrated into the enterprise management
strategy. However, the strategic human resources management activities are not in any
situation can effectively play a role in the enterprise human resources strategy is not in
any situation can be effectively implemented. As is known to all, without the support of
employees and investment, strategic enterprise human resource management activities
are difficult to implement. Strategic human resource management is more able to exert
its effectiveness in the context of the employee’s active participation. Although strategic
human resource management literature let us realize the human resources management
into the strategic management of enterprise value, but the integration of human resource
management and the strategy of the enterprise to enhance the study on the mechanism
of enterprise performance is relatively small.
This study first of strategic human resource management promotion mechanism of
enterprise performance are discussed, think three activities of career development and
staff training plan, pay for performance, temporary staff management let employees
to participate in the decision of enterprise management, is a potential regulator of the
relationship between strategic human resource management and enterprise performance.
Then on our country enterprise human resources management activities were investigation,
empirical validation of the theoretical data to obtain the subjective and objective view
of, found strategic human resource management, career development and staff training
plan, pay for performance, temporary staff management will have a positive impact on
enterprise performance. At the same time, career development and staff training plan,
pay for performance, temporary staff management’s positive influence on strategic human
resource management and corporate performance relationship existing utility regulation,
the career development and training plan for employees regulated utility maximization,
and temporary staff management time, pay for performance plan are minimal.
RISTI, N.º E5, 07/2016 69

Acknowledgments
The work of this paper is supported by colleges’ outstanding young talent support
program in Anhui Province (Grant No.2014SQR03); and Suzhou University Bid
invitation project of the integration of urban and Rural Research Center in north Anhui
province (Grant No.SK2015A185). Suzhou University academic technical backbone
project (2016XJGG05).
References
David, F. (1997). Outsourcing the HR function: personal threat or value opportunity.
Strategic Change, 8, 459–468.
Gree, C., Gray, S. (1999). Human resource management outsourcing the make or buy
decision. Academy of Management Executive.13, 85–96.
Hussain, Z., Wallace, J. (2007). The use and impact of human resource information
systems on human resource management professionals. Information &
Management, 44, 74–89.
James, D., Johnson, R. (2013). Human resource metrics and decision support: A
classification framework Original. Human Resource Management Review, 23,
71–83.
Jenster, P. (1999). Deal maker or deal breaker: human resources issues in successful
outsourcing projects. Strategic Change, 8, 263–268.
Nagendra, A., Deshpande, M. (2014). Human Resource Information Systems
(HRIS) in HR Planning and Development in Mid to Large Sized Organizations.
Procedia - Social and Behavioral Sciences, 133, 61–67.
Pereira, C., Ferreira, C. (2015). Identification of IT Value Management Practices and
Resources in COBIT 5. RISTI-Revista Ibérica de Sistemas e Tecnologias de
Stone, D., Deadrick, L. (2015). The influence of technology on the future of human
resource management. Human Resource Management Review, 25, 216–231.
Zafar, H. (2013). Human resource information systems: Information security concerns
for organizations. Human Resource Management Review, 23, 105–113.
70 RISTI, N.º E5, 07/2016

Asymmetric Effect of Energy prices on Energy Saving

and Emission Reduction based on Dynamic panel
GMM Test
Hua Peng 1,2
hi3223222@tom.com
1
College of Economics and Trade, Hunan University, Changsha 410079, China
2
College of Economics and Trade, Hunan University of Commerce, Changsha 410205, China
Pages: 71–83
Abstract: Building the energy price system which reflects the scarcity of resources
under the background of energy saving and emission reduction has important
theoretical and practical significance. In this paper, we can reflect the scarcity of
energy by measuring the shadow price of energy, then based on the data of 35
industrial sectors, we make an test of the asymmetric effect of energy prices on
energy saving and emission reduction through green technological innovation
and industrial structure adjustment. The results show that, energy shadow price
is conducive to promoting energy conservation and emission reduction, but the
different forms of energy shadow price have different effects on energy conservation
and emission reduction. Green technology innovation can be inspired by energy
shadow prices to reduce energy consumption and pollution, but the path blocks
through the “regulation” effect of the industrial structure to promote energy-saving
emission reduction. The effect of the relative price of energy on energy saving and
emission reduction is not significant, and it is Failure to effectively encourage
enterprise green technology research and development and optimization of
industrial structure, which reflects the defect of non-market pricing mechanism.
Keywords: Energy price, energy saving, dynamic panel GMM, asymmetric effect,
empirical analysis
1. Introduction
Energy saving and emission reduction is an important content of “13th Five-Year” plan,
but also the key to the sustainable development of economy China. The development of
China’s industrialization has long relied on the resource type growth path, which takes
the leading position in the industrial growth mode characterized by high investment,
high energy consumption, high emission, low quality, low efficiency and low output.
However, this extensive industrial growth model leads to a large number of resource
consumption and serious environmental pollution, pollution emissions and resource
consumption has been approaching the limit of environmental carrying capacity. China
is already the world’s largest consumer of energy and carbon dioxide emissions, the
international emission reduction pressure is increasing. 2015, Beijing, Tianjin and
RISTI, N.º E5, 07/2016 71

Asymmetric Effect of Energy prices on Energy Saving and Emission Reduction based on Dynamic panel GMM Test
serious fog and haze events highlight the urgency of environmental pollution control,
energy conservation and sustainable development is imminent. November 2015 China
government in the climate conference in Paris on commitment, by 2030, China’s per
unit GDP carbon dioxide emissions compared to 2005 decreased 60% - 65%, and
incorporated into the “45” planning energy-saving binding emission reduction targets.
Therefore, through energy conservation and emission reduction to promote the
transformation of industrial growth mode and sustainable development is the inevitable
way of China’s economic development.
The existing research literature on relationship between energy prices and energy
conservation and emissions reduction mainly focused on energy prices and energy
consumption, energy efficiency and energy prices and the relationship between the
environmental pollution. Energy prices and energy consumption, energy efficiency
aspects, Birol (2000) found through economic means to increase energy prices can
improve energy efficiency and to reduce energy intensity; Fisher-Vanden (2004) from
the effects of the micro level of energy prices. The results showed that energy source
price rise is China’s energy intensity decreased the driving reason, energy prices
contributed to the high rate of 54.4%; Sue (2008) research energy prices rise in the
long term it will induce technological innovation, so as to reduce energy intensity. Fan
(2012) study changes in energy prices and the embodiment of the technical progress
and non-reflected type technical progress of China 33 industry energy intensity
effect, the study found that rising energy prices and its embodiment of technological
progress to reduce the energy intensity of China’s most industries. The results show
that the price of energy relative index by regulating energy efficiency to reduce energy
consumption. Wang (2014) analyzes energy prices on energy efficiency, the results show
that increase energy prices do help to improve the energy efficiency, but only the price
of energy enough high to energy efficiency have a significant impact. Energy prices
and environmental pollution, James (2007) research that changes in energy prices to
encourage energy production activities of the industry less investment while suppressing
the energy put into the production activities in more industries, through the adjustment
and optimization of the economic structure to reduce the pollution of the environment.
Marklund (2007) found that the shadow price mechanism to reduce carbon dioxide
marginal abatement cost, through the economic loss computation reduction pay, on the
use of policies and measures to reduce carbon dioxide emissions and achieve energy
saving and emission reduction. Lin (2013) study found that energy prices distort the
energy efficiency of China to enhance, reduce the energy exploitation and production
of enterprise enthusiasm, thereby increasing environmental pollution, resulting in haze
weather. Leng (2016) study the relationship between energy price distortions and haze
pollution, the results show that energy price distortions have a positive impact on the
haze pollution. The existing relevant literature to clarify the relationship between energy
prices and energy saving and emission reduction provides a useful reference, but there
are three aspects of the problem is worth the improvement and expansion. The energy
prices to measure and relevant literature basically with fuel and power purchase price
index to represent the energy prices, but this measure is difficult to reflect the scarcity of
energy and environmental costs, the existing literature constructed the reflected energy
scarce areas of the energy shadow price, but the paper extends to the industry level, did
72 RISTI, N.º E5, 07/2016

not effectively distinguish and consider the heterogeneity effect on the energy shadow
price (Jie, 2014; Pereira, 2015).
2. Measurement system of energy shadow price

2.1. Research methods
In this paper, we use DEA technology to calculate the energy shadow price to reflect the
scarcity of energy and environmental costs, thus improving the existing energy relative
price system. Shadow price is not the actual market price, but a kind of calculating price.
Energy shadow price is based on the optimal state of energy allocation and reflects the
real value of the energy price, which is conducive to the existing energy price distortion
correction. Assume that the production system has M decision making units (DMU),
each DMU with energy input E and other N X = [ x1 , x2 ,, x N ] inputs, the production
of S kinds of output Y = [ y1 , y2 ,, yS ]. If the production possibility set meets the
boundedness and convexity, the combination of the weak and the combination of the
weak and the strong can be disposed, using the data envelopment analysis (DEA), the
production technology can be modeled as: T = {Y |( E , X )} ∈ R M . T is the production
technology, said to use input E and X to produce Y. The input directed distance function
is expressed as a technical constraint:
µ1γ 1 + µ2γ 2 (1)

Max
s.t. {Y |( E (1 − γ 1 ), X (1 − γ 2 )} ∈ R M (2)
The assumption PY is that the output of the price vector, PE is the energy of the price
vector, PX is the other input price vector. The decision DMU is to maximize the profit
maximization goal:
Max PY Y − PE E − PX X (3)
s.t. D {Y |( E (1 − γ 1 ), X (1 − γ 2 )} = 1 (4)

To solve the problem of maximizing the profit, the Lagrange function is:
Max PY Y − PE E − PX X + λ ( D {Y |( E (1 − γ 1 ), X (1 − γ 2 )} − 1) (5)
Energy can be obtained by solving the first-order conditions relative shadow price is:
PE ∂D {Y |( E (1 − γ 1 ), X (1 − γ 2 )} / ∂E
=− (1 − γ 1 ) (6)
PY ∂D {Y |( E (1 − γ 1 ), X (1 − γ 2 )} / ∂Y
RISTI, N.º E5, 07/2016 73

Assuming that PY = 1 the absolute shadow price of energy can be obtained at this time:
∂D {Y |( E (1 − γ 1 ), X (1 − γ 2 )} / ∂E
PE = − (1 − γ 1 ) (7)
∂D {Y |( E (1 − γ 1 ), X (1 − γ 2 )} / ∂Y
Values γ 1 obtained by solving the linear programming, which λi represents the cross-
M
sectional observation values of the weights, if ∑λ
i =1
i = 1 that variable returns to scale
(VRS), if λi ≥ 0 and remove the weight and constraint of the said returns to scale
invariant (CRS).
Max µ1γ 1 + µ2γ 2 (8)
M M
∑λ Y
i =1
i i ≥ Y ; ∑ λi Ei ≤ E (1 − γ 1 ); (9)
i =1
M M
∑λ X
i =1
i i ≤ X (1 − γ 2 ); ∑ λi = 1;λi ≥ 0 (10)
i =1
2.2. Indicators and data

Related input and output indicators are selected as follows: 1). Labor input, select the
number of Industrial Enterprises above Designated Size to measure the annual average
number of employees; 2). terms of capital investment, the estimation of capital stock
is relatively complex, through the perpetual inventory method to measure capital
stock is the common measure, but the calculation results due to the depreciation rate
and the initial amount of capital is larger, obtained under different assumptions capital
stock differences. In addition, the China industrial economy statistical yearbook does
not provide 2009-2014 industry fixed assets net annual average balance of data, only
industry reports the net value of the fixed asset data, so this chapter to net value of
fixed asset in the end of the year and net value of fixed asset in the end of the year the
average value to replace; 3). In terms of energy inputs, the total energy consumption of
Industrial Enterprises above designated size is measured, and the conversion of standard
coal conversion to 10 million tons of standard coal; 4). in industrial production, because
energy consumption has obvious industrial intermediate inputs properties, this paper
chooses contains intermediate input cost of the total industrial output value, rather than
industrial added value of industrial output.
3. Study design and indicators

3.1. Model setting
According to the research topic, the energy price is divided into two parts: the shadow
price of energy and the relative price of energy, and considering the energy consumption
and environmental pollution:
74 RISTI, N.º E5, 07/2016

ESit | PI it = γ 0 + ϕYit −1 + γ 1 SPit + γ 2 SPMAX it + γ 3 SPRISit + γ 4 SPDESit + γ 5 SPit × ETit + γ 6 SPMAX it × ETit +

γ 7 SPRISit × ETit + γ 8 SPDESit × ETit + γ 9 SPit × ISit + γ 10 SPMAX it × ISit + γ 11 SPRISit × ISit +
γ 12 SPDESit × ISit + γ 13 X it + µi + ε it (11)
ESit | PI it = δ 0 + ϕYit −1 + δ 1 EPit + δ 2 EPMAX it + δ 3 EPRISit + δ 4 EPDESit + δ 5 EPit × ETit + δ 6 EPMAX it × ETit +
δ 7 EPRISit × ETit + δ 8 EPDESit × ETit + δ 9 EPit × ISit + δ 10 EPMAX it × ISit + δ 11 EPRISit × ISit +

δ 12 EPDESit × ISit + δ 13 X it + µi + ε it (12)
Among them, i is the industry, t for the year, µ1 said the industry individual effects, ε it
as a random disturbance. The econometric model (1) is used to test the non-symmetry
effect of energy shadow price to energy conservation and emission reduction, and
the two aspects of energy consumption and environmental pollution are respectively
expressed as the explanatory variables. Considering the continuity and the dynamic of
the explanatory variables, this paper also added a lag phase Yit −1 to construct a dynamic
panel model to test. SPit as shadow price of energy, but also take into account the non-
symmetric test of the need, the shadow price of energy decomposition, the distinction
between energy prices highest sequence SPMAX it , energy prices rose series SPRISit
and energy prices decline SPDESit in three types of sequence. Also for inspection and
energy prices and the green technology innovation, the adjustment of industrial structure
coordination mechanism of energy saving and emission reduction effect, this paper
set shadow price of energy and decomposition and green technology innovation ETit ,
industrial structure adjustment ISit of cross terms, to reflect the indirect impact on
energy saving and emission reduction. X it As control variables, this paper selected
two variables industrial value-added rate and the overall labor productivity measure,
to reflect the influence of industry heterogeneity factors. In addition, environmental
regulation ERit and technological innovation RDit as control variables.
3.2. Index data

The research object of this paper is the 35 industry in China, the time span is 2000-2014
years. The key variables of the indicators are as follows: (1) energy saving and emission
reduction. The energy saving and emission reduction is divided into ES and PI, energy
saving, energy consumption of unit industrial output value of energy consumption to
measure the degree of energy consumption, emission reduction and environmental
pollution comprehensive index to measure. Comprehensive index of environmental
pollution is according to industrial wastewater emissions, industrial waste gas emissions
and industrial solid waste production three aspects, using entropy method of objective
weight are integrated, and can more fully reflect the emissions of industrial pollutants.
(2) Energy prices. The energy price distinction for the energy shadow price (SP) and
energy relative price EP and SP with the second part, based on the DEA estimates the
energy shadow price said EP with the ratio of the energy index and output index of
price said the energy price index with fuel and power prices to construct the index to
replace, output price index for industrial products factory price index to replace and
RISTI, N.º E5, 07/2016 75

are converted as the base year unchanged price index. In addition, the energy shadow
price and energy relative price are decomposed into the highest sequence PMAX, rising
sequence PRIS and descent sequence PDES, the decomposition formula of the three
kinds of sequences is:
Pit = PMAX it + PRISit + PDESit , i = 1, 2, 3 N ,t = 1, 2, 3T (13)

= =
PMAX it MAX ( Pi0 , Pi1 , Pit ),t 1, 2, 3T (14)

t
PDESit = ∑ MIN {0,(PMAX ij −1 − Pij-1 ) − ( PMAX ij − Pij )},t = 1, 2, 3T (15)
j =1
The selection of intermediate variables and control variables are as follows: (1) green
technological innovation. Et considering undesirable outputs and based on SBM
model productivity indicator to measure input and output index choice of labor input,
capital investment, energy input and total industrial output, non-industrial CO2, SO2
and industrial cod selection of desired outputs. (2) industrial structure adjustment. IS
with the state-owned and state holding industrial enterprises accounted for more than
the total assets of industrial enterprises accounted for the proportion of total assets to
measure the impact of industrial property structure adjustment on energy saving and
emission reduction. (3) Dependent control variables. Industrial added value rate of RV
for the industrial added value and total industrial output value over the same period
ratio, overall labor productivity of LP for the industrial increase value and the average
number of staff and workers ratio, environmental regulation by the industry wastewater
and waste gas treatment operating costs and industrial gross output value of the ratio
measure, scientific and technological innovation in the industry science and technology
activities expenditures total accounted for the proportion of total industrial output value
to measure. All related data are derived from China’s industrial economic statistical
yearbook, China Environmental Statistics Yearbook, China Statistical Yearbook of
science and technology, and statistical yearbook of China.
4. Asymmetric effect test

4.1. Panel unit root and co integration test
This paper uses four methods of LLC test, IPS test, Fisher-ADF test and Fisher-PP test
to test the relative variables in the panel unit root test. Test results show that all the
variables of the level sequence are not in the four statistical tests at the same time rejected
the original hypothesis, indicating that the original sequence is not smooth. And all of
the variables of the first order differential sequence are at the level of 1%, rejected four
statistical tests of the original hypothesis, which shows that the first order difference is
smooth. Therefore, all of the variables of this paper are first order differential smooth I
(1), which can be tested by panel co integration test.
76 RISTI, N.º E5, 07/2016

Horizontal sequence statistics First order difference statistics

variable
LLC IPS ADF PP LLC IPS ADF PP
-2.735 3.235 45.612 50.720 -18.322 -10.17 216.82 289.35
ES
(0.003) (0.999) (0.988) (0.962) (0.000) (0.000) (0.000) (0.000)
explained variables
6.285 11.535 2.468 2.275 -9.358 -3.576 108.522 127.622
PI
(1.000) (1.000) (1.000) (1.000) (0.000) (0.000) (0.002) (0.000)
-3.126 2.096 60.536 67.215 -21.256 -11.675 240.123 270.998
SP
(0.002) (0.986) (0.785) (0.584) (0.000) (0.000) (0.000) (0.000)
explanatory variables
-29.075 -12.615 27.088 12.323 -35.535 -40.286 49.935 61.123
SPMAX
(0.000) (0.000) (0.007) (0.418) (0.000) (0.000) (0.000) (0.000)
-0.225 3.858 27.936 27.566 -7.443 -3.322 102.276 189.853
SPRIS
(0.358) (0.999) (0.944) (0.988) (0.000) (0.000) (0.000) (0.000)
-1.522 4.068 37.538 53.786 -18.224 -10.680 229.635 310.145
SPDES
(0.063) (1.000) (0.999) (0.935) (0.000) (0.000) (0.000) (0.000)
2.122 6.848 16.566 19.315 -23.862 -14.582 288.968 454.265
EP
(0.958) (1.000) (1.000) (1.000) (0.000) (0.000) (0.000) (0.000)
0.825 0.274 52.338 85.360 -17.572 -8.195 170.105 196.745
EPMAX
(0.795) (0.608) (0.750) (0.018) (0.000) (0.000) (0.000) (0.000)
-6.295 2.768 33.672 39.143 -13.486 -5.778 147.872 345.426
EPRIS
(0.000) (0.998) (0.999) (0.999) (0.000) (0.000) (0.000) (0.000)
-5.722 2.025 37.158 30.770 -17.442 -7.163 163.426 258.052
EPDES
(0.000) (0.988) (0.999) (1.000) (0.000) (0.000) (0.000) (0.000)
13.018 -1.792 82.588 253.116 -18.885 -15.122 299.686 527.226
ET
(1.000) (0.036) (0.145) (0.000) (0.000) (0.000) (0.000) (0.000)
-4.505 4.770 42.035 29.126 -16.536 -9.068 196.772 288.786
IS
(0.000) (1.000) (0.996) (1.000) (0.000) (0.000) (0.000) (0.000)
-0.032 -2.115 99.854 230.470 -17.852 -13.882 280.692 538.927
ER
(0.488) (0.018) (0.012) (0.000) (0.000) (0.000) (0.000) (0.000)
4.308 -0.116 47.578 98.035 -15.575 -7.368 175.105 361.828
RD
(1.000) (0.454) (0.984) (0.015) (0.000) (0.000) (0.000) (0.000)
control variable
-3.112 1.530 56.348 63.958 -11.192 -8.488 188.922 318.315
RV
(0.000) (0.936) (0.882) (0.680) (0.000) (0.000) (0.000) (0.000)
4.345 7.062 39.812 41.698 -9.235 -3.433 117.434 156.733
LP
(1.000) (1.000) (0.998) (0.998) (0.000) (0.000) (0.000) (0.000)
Table 1 – Panel unit root test

This paper uses panel co-integration test based on E-G two-step test pedroni and Kao
test, test results show that, although the rho-Statistic Panel and rho-Statistic Group
two statistics did not pass the significant test, but the other five statistics are at 5% levels
to reject the original hypothesis does not exist co integration relationship. Considering to
the sample interval belongs to the small sample, combined with the panel ADF-Statistic
and group ADF-Statistic shows that there exist co-integration relationship and Kao test
results also significantly declined the null hypothesis, show that there is a co integration
relationship between variables, energy prices on energy saving and emission reduction
of measurement estimation.
RISTI, N.º E5, 07/2016 77

Correlation statistics Statistic value Pvalue

Panel v-Statistic 2.665 0.003
Panel rho-Statistic 2.838 0.968
Panel PP-Statistic -4.935 0.000
Panel ADF-Statistic -4.008 0.000
Group rho-Statistic 5.427 1.000
Group PP-Statistic -9.836 0.000
Group ADF-Statistic -7.898 0.000
Kao ADF-Statistic -1.936 0.025
Table 2 – Panel co-integration test
4.2. Asymmetric effect estimates

In this paper, the dynamic panel system GMM estimation method is used to deal with
the problem caused by the lag phase of the explained variable, to test the non symmetry
effect of energy price to energy saving and emission reduction. It was found that the AR
(1) test and AR (2) test showed that there was a first order sequence correlation, but there
was no correlation of the two order sequences. At the same time, Sargan test results also
show that the tool variables selected in this paper are effective, and the measurement
model is reasonable. The coefficients of Y-1 are all over 0.8, and they are all significant
at 1% level, which indicates that the energy consumption and environmental pollution
have strong dynamic and continuity. In view of the large estimation coefficient of Y-1,
this paper uses the system GMM estimation method, which is more robust than the
GMM estimation method.Asymmetric effects of energy prices on energy consumption
estimation results show that the energy shadow price (SP) on the energy consumption
strength es have negative effects, and the estimated coefficient is significant, indicating
that the energy shadow price rise can reduce energy consumption intensity.
Energy shadow price relative price of energy
variable
Model 1 Model 2 Model 3 Model 4
Y-1 0.922***(110.98) 0.886***(494.80) 0.895***(57.35) 0.828***(57.25)
SP -26.538**(-2.40) -10.045***(-2.82)
SPMAX -26.516**(-2.42) -10.066***(-2.80)
SPRIS -26.535**(-2.40) -0.606***(-10.24)
SPDES 26.538**(2.40) 0.628***(10.25)
SP×ET -392.688*(-1.95)
SPMAX×ET -392.338*(-1.96)
SPRIS×ET -390.132*(-1.95)
SPDES×ET 390.286*(1.95)
SP×IS 0.608***(10.24)
SPMAX×IS 0.609**(10.25)
78 RISTI, N.º E5, 07/2016

SPRIS×IS 0.603***(10.23)
SPDES×IS 10.122***(2.82)
EP 0.862(0.32) 0.070(0.75)
EPMAX -0.895(-0.34) -0.068(-0.88)
EPRIS -0.898(-0.34) -0.065(-0.35)
EPDES 0.224***(5.58) 0.004**(2.28)
EP×ET 30.465(0.55)
EPMAX×ET -30.995(-0.45)
EPRIS×ET -25.215(-0.35)
EPDES×ET 0.006***(57.60)
EP×IS 0.576(0.26)
EPMAX×IS -0.572(-0.23)
EPRIS×IS -0.588(-0.24)
EPDES×IS 0.004**(2.24)
RV -1.512***(-7.68) -0.002**(-2.38) -0.009***(-7.78) -0.002***(-10.27)
LP -0.004*(-1.76) 0.002(0.38) 0.004(1.35) 0.148(1.42)
ER -0.003***(-3.45) -0.005***(-5.24) -0.003***(-3.28) -0.002***(-2.62)
RD -0.055***(-12.34) -0.038***(-5.25) -0.035***(-6.25) -0.025**(-2.30)
AR(1) test -2.766[0.005] -2.598[0.009] -2.608[0.009] -4.484[0.000]
AR(2) test 0.458[0.650] 0.588[0.555] 0.302[0.762] 1.578[0.115]
Sargan test 27.710[1.000] 27.522[1.000] 26.886[1.000] 34.220[1.000]
Number of samples 490 490 490 490
Table 3 – Asymmetric effect of Energy price to energy consumption ES

By observing the energy shadow price (SP) through the innovation of green technology
et on energy consumption strength of ES indirect effect shows, model 1 energy shadow
price (SP) and green technology innovation et cross term SP * et coefficient is significantly
negative, and combined with the estimated coefficients of model 1 energy shadow price
(SP), is not difficult to draw the energy shadow price can be through incentives for
green technology innovation to reduce the energy consumption of the conclusion of
the study. Actually, when the most energy Unisys, rise in energy prices will inevitably
increase the cost of production of industrial enterprises, in the long run will inspire
enterprises through green technology research and development or factor substitution
to reduce costs, to continue to maintain the core competitive advantage. And, the energy
shadow price highest sequence SPMAX and rising sequence spris cross coefficients
are consistent with the SP * et and also verify the energy shadow price rise and green
technology innovation synergy will help to reduce the energy consumption intensity.
But the energy shadow price decreased sequence SPDES and endothelin (ET) of cross
terms SPDES x et coefficient is significant for is, highly consistent with the estimated
coefficients of SPDES, that even if the energy in the optimal use of state, the decline
in energy prices still enthusiasm may reduce industrial enterprises to carry out green
technology research and development, in the long run is not conducive to technological
RISTI, N.º E5, 07/2016 79

progress and energy conservation, once again proved the existence of asymmetric effect.
Study found that energy prices down to the technical efficiency improvement of neglect
and price recovery period of inefficient and irrational will make energy consumption
into a vicious circle, different cycle fluctuations in energy prices in the regulation of
energy technical efficiency control energy consumption exist significant “asymmetric”
moderating effect. In addition, this paper used to reflect the energy scarcity of energy
shadow prices to conduct an investigation; the estimated results are relatively robust.
Relative energy price EP on energy consumption es estimation results show that, the
estimated coefficients of model 3 and model 4 energy shadow price EP is positive, the
highest sequence EPMAX and rising estimated coefficients of sequence EPRIS negative,
but were not significant.
Energy shadow price relative price of energy

variable
Model 1 Model 2 Model 1
Y-1 0.882***(86.45) 0.858***(36.44) 0.838***(44.75) 0.866***(138.25)
SP -3.768**(-2.28) -0.004**(-2.22)
SPMAX -3.855**(-2.34) -0.003**(-2.20)
SPRIS -3.728**(-2.25) -0.005**(-2.28)
SPDES 3.845**(2.34) 0.005**(2.25)
SP×ET -0.178***(-5.52)
SPMAX×ET -0.174***(-5.62)
SPRIS×ET -0.175***(-5.56)
SPDES×ET 3.824**(2.36)
SP×IS 0.002***(4.12)
SPMAX×IS 0.003***(5.15)
SPRIS×IS 0.005***(12.18)
SPDES×IS 0.248***(6.68)
EP -0.002(-0.33) -0.012(-0.88)
EPMAX -0.003(-0.34) -0.014(-0.90)
EPRIS -0.008(-0.85) -0.018(-1.28)
EPDES 0.002***(6.87) 0.016***(3.96)
EP×ET -0.488(1.62)
EPMAX×ET -0.485(-1.62)
EPRIS×ET -0.426(-1.42)
EPDES×ET 0.216***(34.58)
EP×IS 0.002(0.18)
EPMAX×IS 0.002(0.20)
EPRIS×IS 0.002(0.29)
EPDES×IS 0.018***(3.76)
RV -0.008***(-7.78) -0.008***(-7.70) -0.002***(-10.62) -0.002***(-9.52)
80 RISTI, N.º E5, 07/2016

LP -0.005***(-4.75) 0.005(0.75) -0.002(-0.55) 0.003(0.48)

ER -0.005***(-11.42) -0.006***(-11.45) -0.002**(-2.52) -0.002***(-4.78)
RD -0.038***(-39.80) -0.032***(-39.82) -0.007***(-9.37) -0.006***(-8.35)
AR(1) test -2.580[0.010] -4.125[0.000] -4.578[0.000] -4.560[0.000]
AR(2) test 0.638[0.522] 1.420[0.155] 0.855[0.400] 0.982[0.322]
Sargan test 26.956[1.000] 33.618[1.000] 33.924[1.000] 34.022[1.000]
Number of samples 490 490 490 490
Table 4 – Asymmetric effect of Energy prices on environmental pollution PI

In addition, table 4 shows the results of the non-symmetric effect of energy price on
environmental pollution. The results are consistent with the table 3. AR (1) test, AR (2)
test and Sargan test show that the model is reasonable and the instrumental variables
are effective. Energy shadow price (SP), the highest sequence SPMAX and sequence
spris rises to the pollution of the environment has obvious inhibitory effect, but a
decreasing sequence SPDES but has positive effects, shows that the energy shadow
price to the pollution of the environment also has asymmetric effects. Energy shadow price
(SP) mainly through the ET of green technology innovation to curb energy consumption
intensity and energy shadow price (SP), the highest sequence SPMAX, rising spris
sequences and et cross coefficient is significantly negative, provide good support for
the hypothesis. However, the decline in energy shadow prices will lead to excessive and
inefficient use of energy, thereby increasing environmental pollution emissions. Energy
shadow price SP also failed to achieve pollution reduction by adjusting the industrial
structure of IS, the combination of the two instead of increasing environmental pollution
emissions. Control variables, in addition to total labor productivity LP on energy
conservation and emission reduction effect is not significant, the industrial value-added
rate, environmental regulation, and technological innovation are conducive to energy-
saving emission reduction. Especially technological innovation is of great importance to
energy-saving emission reduction, not only can direct inhibition of energy consumption
and environmental pollution by means of energy-saving emission reduction technology,
more important is conducive to play the effects of regulation of energy prices, to cooperate
to promote the realization of energy-saving emission reduction effect.
5. Conclusions and policy implications

The conclusion of this paper shows that the existing non-market pricing mechanism
in China has become a great obstacle to the further development of energy saving and
emission reduction, and the reform of energy pricing mechanism is imperative. In
addition, the unreasonable industrial structure is not conducive to the full play of the
energy price adjustment effect, based on the conclusions of this paper put forward to
promote energy-saving emission reduction policy implications: 1).The energy market
price system which reflects the scarcity of energy should be constructed effectively.
Reflected in the shadow price of energy is energy optimal use of the marginal value,
which is the important information of industrial enterprises determine the amount
of energy input, when energy prices relatively low industrial enterprises will choice
RISTI, N.º E5, 07/2016 81

more energy investment, which leads to low efficiency of energy utilization. Therefore,
it is the key to ease the pressure of energy and promote the energy conservation and
emission reduction to build the energy market price system which can reflect the
shadow price of energy. In addition, rising energy prices is an effective way to reduce
energy consumption and environmental pollution, improve energy efficiency; 2)
Through the financing support, financial subsidies and other ways to encourage non-
state-owned economy into the energy consuming and polluting industries, in particular,
to attract advanced pollution control technologies and cleaner production technology
and have better research and development capabilities of enterprises to enter. But
the government cannot go into the blind implementation of the industrial structure
adjustment, and to use the difference of the adjustment strategy. Of nationalization and
privatization ratio relatively reasonable area, increase the focus on the introduction of
private scientific and technological innovation of enterprises and personnel support and
collocation other incentives to guide enterprises to carry out green technology research
and development; area of state-owned share is too high or privatization proportion is
too low, moderately reduce nationalization rate or increase the private rate based, but
also increase the innovation of science and technology and talent introduction policy
support. Only by reforming the market energy price system and encouraging the green
technology innovation and adjusting the industrial structure, can we make full energy
price adjustment effect to promote energy-saving and emission reduction.
References
Birol, F., Keppler, J. (2000). Prices, Technology Development and the Rebound Effect.
Energy Policy, 28, 457–469.
Fan, M. (2012). Energy prices, technological change and the impact of information
technology investment on the energy intensity of the sector. world economy, (5),
22–45.
Fisher-Vanden, K., Jefferson, H. (2004). What is Driving China’s Decline in Energy
Intensity? Resource and Energy Economics, 26, 77–97.
James, B. (2007). CO2 Emissions,Energy Consumption, and Output in France. Energy
Policy, 35 (10), 4772–4778.
Jie, C. (2014). The moderating effect of internal energy price relative index on energy
consumption: a study based on the perspective of technical efficiency change.
Resource science, (3), 520–529.
Leng, L., Du, S. (2016). Energy price distortion and fog and haze pollution - China’s
empirical evidence. Industry economic research, (1), 71–79.
Lin, B., Du, K. (2013). Influence of factor market distortion on energy efficiency.
economic research, (9),125–136.
Marklund, O., Samakovlis, E. (2007). What is driving the EU Burden-sharing Agreement:
Efficiency or Equity? Journal of Environmental Management, 85(2), 317–329.
82 RISTI, N.º E5, 07/2016


Sue, W. (2008). Explaining the Declining Energy Intensity of the U.S. Economy.
Resource and Energy Economics, 30, 21–49.
Wang, J. (2014). The impact of energy prices on energy efficiency - Empirical Analysis
Based on global data. Economic management, (12), 13–23.
RISTI, N.º E5, 07/2016 83

BIRCH Algorithm and Data Mining Application in

Construction of Financial Enterprise Management
Team based on “Internet Plus” Background
Chenglin Xiao, Weili Xia*
* xiaweili@nwpu.edu.cn
Northwestern Polytechnical University, Xi ‘an 710072, Shaanxi, China

Pages: 84–94
Abstract: According to a series of problems in the current situation of financial

management team under Internet background, the authors construct the index system of
teammanagementbasedondataminingmethod.Theresultsshowthatthefourdimensions
X2/df=4.287<5, RMSEA=0.063<0.08, GFI=0.95,CFI=0.94,IFI=0.94,NNFI=0.93.
Built element factor analysis, KMO=0.826, Barlett test value P=0.000<0.01, the total
variance of the 4 factors explain the cumulative as 70.911%.The main conclusions is
Chinese enterprise financial management team consists of four dimensions, respectively
as team base, team cohesion and decision making, team goals and team effectiveness,
and there is a high correlation between them.
Keywords: Data mining, Internet plus, financial enterprise, management team,
construction elements
1. Introduction
Global economic integration has become the main development trend in the world,
with China market economy and network economy has gradually become mature
and norms, “Internet plus” represents a new form of social economy, give full play to
the optimization of the Internet in the allocation of social resources and integration
for the whole society to enhance innovation and productivity, the formation of more
widely the Internet infrastructure development and implementation of the new form
of economic tools (Vicky, 2015; Pereira, 2015; Mei-Hsiang, 2016).Team management
concept in foreign research started earlier, and successfully applied to various forms
of organization and attempt to apply it to their own management practice, with a view
to improve management effectiveness, improve management efficiency to provide
meaningful reference and help (Julia, 2013; Dara,2012). Domestic enterprises is the
first to learn and use the organization of the team work, but the development of its team
building and management model, in general, is still relatively backward. The research
content involves the formation of the team, the team goals, team decision-making,
team cohesion, team relationships, team learning and team performance (Dara, 2013;
Wonjung, 2015; Belinda, 2015). According to the domestic and foreign research status
of enterprise management team, on the basis of the experience of mature management
84 RISTI, N.º E5, 07/2016

system at home and abroad on the “Internet plus” under the background of Chinese
financial enterprise management team construction and development.
2. BIRCH Algorithm and Data Mining

2.1. BIRCH Algorithm
BIRCH algorithm itself is a clustering algorithm, but it overcomes some of the
shortcomings of the K-Means algorithm, such as the determination of the K, because
the algorithm itself is not set before the number of clusters. He is achieved through
the CF-Tree, (Cluster Feature-Tree) clustering feature tree. An important consideration
of BIRCH is to minimize the I/O, by scanning the database, the establishment of a
stored in the memory of the initial CF- tree, can be seen as a multi data compression.
Figure 1 –BIRCH Algorithm Process

Cluster analysis based on similarity of data objects grouped; found that the distribution
of data space characteristics is a kind of data mining method. Hierarchical clustering
method can be divided into hierarchical clustering and hierarchical clustering.
Hierarchical clustering uses a bottom-up strategy for clustering, which starts from
the single member clustering, and gradually merges them into a larger cluster. In
each layer, the nearest two clusters are merged. On the contrary, it is the split level
clustering. BIRCH algorithm combined with the hierarchical aggregation and iterative
re positioning method, first with the bottom-up hierarchical algorithm, and then uses
the iterative relocation to improve the results. The main idea is to scan the database
and build an initial cluster feature tree which is stored in the memory, and then cluster
the leaf nodes of the tree. BIRCH algorithm is a very effective, traditional hierarchical
clustering algorithm; the algorithm can be used to scan the scan efficiently and effectively
to deal with outliers.
RISTI, N.º E5, 07/2016 85

BIRCH Algorithm and Data Mining Application in Construction of Financial Enterprise Management
Figure 2 – Clustering feature vector
2.2.Data mining
With the rapid development of information technology, many industries, such as
commercial, enterprise, scientific research institutions and government departments have
accumulated massive, different forms of data storage. These data often implies all sorts of
useful information, rely solely on database query retrieval mechanism and statistics method
is difficult to obtain the information, an urgent need to automatically and intelligently to
be processing of data into valuable information, so as to achieve for the purpose of service
decision. In this case, a new technology - Data Mining (DM) technology came into being.
Data mining is a multi-disciplinary field, it combines the database technology, artificial
intelligence, information retrieval and other research results of the latest technology, and
its application is very extensive. As long as there is an analysis of the value of the database,
you can use the data mining tools to dig useful information.
Figure 3 – Data mining
86 RISTI, N.º E5, 07/2016

3. Research object and methods

3.1. Research object
According to the Chinese enterprise management in Colleges and universities of visibility
and geographical distribution, out of six colleges and universities (Northwestern
Polytechnical University, Peking University, Renmin University of China, Zhejiang
University, Fudan University, Xiamen University) 10 experts were interviewed, and to
them for special interview personnel to study object, table 1 shows.
index classification number Proportion (%)
male 7 70.00
Gender
female 3 30.00
Under 35 years of age 1 10.00
36-45 years old 3 30.00

Age
46-55 years old 4 40.00
Over 56 years of age 2 20.00
doctor 4 40.00
academic degree master 3 30.00
bachelor 3 30.00
professor 3 30.00
title associate professor 6 60.00
lecturer 1 10.00
Northwestern Polytechnical University 3 30.00
Peking University 2 20.00
Renmin University of China 2 20.00

school
Zhejiang University 1 10.00
Fudan University 1 10.00
Xiamen University 1 10.00
Table 1 – Interviewers basic characteristic information (n=10)

According to Chinese geographical rational distribution, remove 10 financial
enterprises (Bank of China Erenhot City branch, Bank of Inner Mongolia Xilingol
branch, The Industrial and Commercial Bank of China, Head Office, Xi’an Branch
of China Minsheng Banking Corp., Ltd., Bank of China Xi’an, Chengdu Branch of
China Minsheng Banking Corp., Ltd., Ningbo Yinzhou Rural Cooperative Bank,
Bank of Jiangxi, Jiujiang bank, Hankou bank) of 316 enterprise management team is
questionnaire survey, and to them as positive test questionnaire researchers object.
Table 2 results show that result.
RISTI, N.º E5, 07/2016 87

index classification number Proportion (%)

male 229 72.47
Gender
female 87 27.53
35 years old and below 83 26.27
36-45 years old 166 52.53
Age
46-55 years old 45 14.24
56 years old and above 22 6.96
Department 3 0.95
rank of section chief 45 14.24
level
administrative rank of section or department 123 38.92
ordinary member of 145 45.89
5 years and below 115 36.39
11 - 6 years 85 26.90
Working life
12-18 years 77 24.37
19 years and above 39 12.34
doctorate 12 3.80
master’s degree 57 18.04
academic degree
Bachelor’s degree 168 53.16
nothing 79 25.00
Table 2 – Basic characteristic information of the questionnaire (n=316)
3.2. Research methods

1. Document method: The article as much as possible fully searchable research
results at home and abroad, absorption and digestion of domestic and foreign
related enterprise management team literature, understand the frontier and
the progress of the theory of enterprise management team, especially scientific
theories and methods used by foreign scholars, from their existing research results
and the method, put forward the theoretical framework and hypotheses, and the
empirical research in the concept of measurement data sheet for a useful reference.
2. Interview method: Interview is mainly conception of this study are
validated and the further adjustment, and as a measure of the conceptual model
and research the variable table determine specific content and experience of
the experts and the interview of the relevant authority of the experts for the
questionnaire lay a solid foundation.
3. Questionnaire survey: Questionnaire survey is an important method
for the research of this study, through the discussion on the structure of the
empirical study on variables measuring scale and in formal empirical research
by larger scale questionnaire validation study ideas and research hypothesis,
for the construction of the article and the results provide security. All of the
questionnaires were evaluated by the method of grade five. Score for the 10-2
points, very important for 10 points, more important for 8 points, the general
88 RISTI, N.º E5, 07/2016

important for 6 points, compared with 4 points, is not important for 2 points,
the higher the score, the higher the degree of importance.
4. Mathematical statistics method: Mathematical statistics is mainly used to
determine the value of the collected data, to verify the theoretical assumptions.
According to the needs of the research purposes, this study uses statistical
software package SPSS16.0, EXCELL and LISREL8.51 three kinds of software.
5. Logic inference method: By using the methods of comparison, deduction,
induction, analysis and reasoning, the results of the statistical significance of the
data are analyzed, and the corresponding theoretical results are obtained.
4.1. Survey data
Study of 10 expert’s interview, mainly take the following two ways: one is carries on
the induction and the summary of previous enterprise team management content and
measurement; the second is semi-structured interviews were conducted. The results were
recorded, analyzed and summarized. The researchers prepared according to interviews of
some of the indicators. At the same time, interviews were carried out frequency analysis
and ranking, and the enterprise team management content 22 see table 3.
Content index Fre. Per.% Ran. Content index Fre. Per.% Ran.
1 people’s Academic 2 people’s work
9 90 4 7 70 18
Degree Foundation management life
4 person’s
3 basic skills of language
8 80 10 comprehensive quality 8 80 10
expression
foundation
5 the total number of 6 team development
10 100 1 7 70 18
teams time
8 team members’
7 team culture 9 90 4 8 80 10
harmony and trust
9 communication and
10 work together as a
collaboration of team 8 80 10 9 90 4
team
members
12 leader decision
11 team work spirit 7 70 18 10 100 1
making ability
13 public decision-making 14 crisis decision-
8 80 10 10 100 1
capacity making ability
15 collective goals 9 90 4 16 recent target 8 80 10
18 medium and long
17 personal goals 8 80 10 9 90 4
term goals
19 meet customer needs 9 90 4 20 team growth 7 70 18
21 to meet the needs of the
8 80 10 22 team value 7 70 18
players
Table 3 – Financial enterprise group management content ranking (n=10)
RISTI, N.º E5, 07/2016 89

4.2. Test analysis

35 questionnaires were issued to 10 financial enterprises, a total of 350 copies, 335
copies were recovered, the recovery rate was 95.71%, the effective questionnaires were
316, and the effective rate was 94.33%. Using SPSS16.0 to exploratory factor analysis
of the results of the pilot survey, using the principal component method to extract the
common factor, and take the characteristic value of more than 2 to intercept the standard
of the common factor. In “the enterprise team management” 16 related projects, table 4
results show that four factors explain the total variance for total 70.911%, which holds
the public factor occupancy 26.496%, the male factor 17.504%, the third common
factor occupies 13.872% and the four public factor occupies 13.039%). For the study,
the results of table 5 show that the common factor 1 is named as the team “based factor
of the enterprise team management”, Common factor 2 named “human team cohesion
and decision ability factor”, common factor 3 is named the “team goal factor of the
enterprise team management”, Common factor 4 named “team effectiveness factor for
enterprise team management”.
Initial Eigenvalues Rotation Sums of Squared Loadings

Component % of
Total Cumulative % Total % of Variance Cumulative %
Variance
1 6.103 27.741 27.741 5.829 26.496 26.496
2 4.186 19.029 46.770 3.851 17.504 44.000
3 2.981 13.551 60.321 3.052 13.872 57.872
4 2.330 10.589 70.911 2.869 13.039 70.911
Table 4 – Total Variance Explained(n=316)
Component
index
1 2 3 4
4 people’s Academic Degree Foundation 0.880
6 people’s work management life 0.839
10 person’s language expression ability foundation 0.912
12 person’s comprehensive quality foundation 0.871
15 the total number of teams 0.631
17 team development time 0.666
18 team culture 0.709
1 team members’ harmony and trust 0.566
7 communication and collaboration of team members 0.680
8 work together as a team 0.800
17 team work spirit 0.842
90 RISTI, N.º E5, 07/2016

Component
index
1 2 3 4
18 leader decision making ability 0.576
19 public decision-making capacity 0.842
20 crisis decision-making ability 0.621
5 collective goals 0.709
9 recent target 0.505
14 personal goals 0.633
16 medium and long term goals 0.709
2 to meet customer needs 0.501
3 team growth 0.618
11 to meet the needs of the players 0.552
13 team value 0.793
Table 5 – Rotated Component Matrix(n=316)
4.3. The reliability and validity test

Reliability analysis is used to analyze the stability and the consistency of the data, the
general use Cronbach’s alpha coefficient was used to observe internal consistency, in
the questionnaire, a coefficient >0.7, that data stability and consistent performance
meet the research needs, and the data has a good reliability; validity analysis is mainly
observed the design index of validity and rationality, commonly used Bartlett ball
degree (Barlett) index to analyze, think statistically observations is relatively large
and the p value less than the significant level, and shown to do factor analysis. Kmo
test comparison of simple correlation coefficient between the variables and partial
correlation coefficient of index, index between 0 - 1, closer to 1, the better was that
KMO>0.9 is very good and very suitable for, KMO>0.8 said better and is more suitable
for, KMO>0.7 said, KMO>0.6 said less and less suitable. KMO<0.6 said very well
and is not suitable. Table 6 the results show that the enterprise team management
questionnaire reliability and validity of the test results of the questionnaire, Cronbach
a=0.873>0.7, that data stability and consistent performance meet the research needs,
the data has good reliability; kmo value =0.826 were > 0.8 that better and more suitable
for; and Barlett sphericity test p value = 0.000, statistics to have very significant
difference in meaning, that the structural validity of the questionnaire is very good.
Scale items number Cronbach a KMO P value

Enterprise team management scale 38 0.873 0.826 0.000
Table 6 – Questionnaire reliability and validity test (n=316)

Reasonable assumptions verification “enterprise team management” four dimension
structure, raise exceptions three competition model to carry on the comparison,
RISTI, N.º E5, 07/2016 91

respectively, for dimension 3, 2 and 1 dimension, of which 3 dimensions including (“team

based”, “team cohesion and decision ability and team goals and team effectiveness”
merger); 2 dimensions including (“team based” and with “team cohesion and decision
ability and team goals and team effectiveness” merger); 1 dimensions the four factors
all combined. Table 7 results show: “enterprise team management” four dimensional
structure to better sample, the fourth dimension X2/df=4.287<5, RMSEA=0.063<0.08,
GFI=0.95, CFI=0.94, IFI=0.94, NNFI=0.93 are greater than 0.90. Visible, four
dimensional model fit the ideal, compared with the other 3 dimensional model, the
four dimensional model of the fitting data index is obviously better than the other 3
dimensional model, which shows that the hypothesis is reasonable.
Model X2 df X2/df GFI RMSEA RMR NFI NNFI CFI IFI

4 dimension 372.969 87 4.287 0.95 0.063 0.065 0.94 0.93 0.94 0.94
4 dimension 456.926 89 5.134 0.86 0.11 0.087 0.83 0.82 0.79 0.81
2dimension 597.223 91 6.563 0.75 0.18 0.094 0.75 0.74 0.72 0.72
1 dimension 758.694 93 8.158 0.68 0.22 0.099 0.67 0.66 0.70 0.70
Table 7 – Comparative analysis of confirmatory factor results(n=316)
4.4. Interaction of each dimension

Table 8 results show: the enterprise team management index content, team based, team
cohesion and decision-making ability, team goals, team effectiveness of correlation
coefficient respectively 1, 0.81, 0.67 and 0.561, moderate, visible, “the enterprise team
management index does not exist independently and internal mutual has moderate or
related. “Team based” refers to a formal group element that is composed of individuals
who work together to achieve a certain goal. Must have: one is groups emphasize
information sharing team stressed the collective performance; second, the role of group
is neutral (sometimes negative), and the role of the team is often positive; three is a
group of individual liability, and responsibility to the team may is an individual, may
also be common; four skill groups is random or different, and team skills is complement
each other.
Serial content 1 2 3 4
1 Team base 1
2 Team cohesion and decision making 0.81 1
3 Team goals 0.67 0.55 1
4 Team effectiveness 0.61 0.59 0.77 1
Table 8 – Correlation matrix analysis table (n=316)
92 RISTI, N.º E5, 07/2016

5. Conclusions
The team management of commercial banks is the foundation to achieve business success.
Excellent commercial bank team is a constant pursuit of excellence, innovation, and
complementary skills, willing to work together for the common purpose and performance
objectives and methods and to work together for the team. Commercial bank itself is a big
team, under the various functional departments, grass-roots units are commercial banks
in the sub team. How to scientifically and effectively to these good management team,
to create the best performance is the important objectives of management team. A good
team must highlight employee’s personal values, respect the status of employees in the
team, to provide staff development space, the staff of professional value realization and the
enterprise’s survival and development requirements are organically combined together.
In this team, employees want mutual help and support requirements, in a collective
work, promote cooperation between staff and improve staff work of excellence “sense of
achievement”, and create a increased job satisfaction of the atmosphere of mutual trust
and the vast space in the team. This is the team spirit, is the ideological core of a good team.
Performance evaluation through the control completion of work objectives, adopt a scientific
approach and assessing the employee’s completion of work objectives, responsibilities to
fulfill degree. The purpose of performance appraisal is to improve the performance of the
team staff, and to link their work and the development goals of the enterprise. Effective
performance evaluation is an important means to improve the efficiency and management
efficiency of enterprise management. So that only scientific, practical work performance of
team workers were evaluated on the basis of the establishment of effective incentive and
remuneration system, in order to give full play to the employees play an important role in
the process to achieve business goals and in the fierce competition in the human resources
made active, so that commercial banks really have a branch in the competition remain
invincible position in the talent team.
References
Mei-Hsiang, W., Tarng, Y. (2016). Investigating the success of knowledge management:
An empirical study of small- and medium-sized enterprises. Asia Pacific
Management Review, 21, 79–91.
Vicky, A., Tanya, B. (2015). Leveraging integrated information systems to enhance
strategic flexibility and performance: The enabling role of enterprise risk
management. International Journal of Accounting Information Systems, 19, 1–16.
Julia, E., Hoch, J. (2013). Shared leadership in enterprise resource planning and human
resource management system implementation. Human Resource Management
Review, 23, 114–125.
Dara, P., Sari, A. (2012). Corporate Governance Mechanism and the Level of Internet
Financial Reporting: Evidence from Indonesian Companies. Procedia Economics
and Finance, 2, 157–166.
RISTI, N.º E5, 07/2016 93

Wonjung, N., Ji, Y. (2015). Nurses’ Educational Needs Assessment for Financial
Management Education Using the Nominal Group Technique. Asian Nursing
Research, 9, 152–157.
Belinda, S. (2015). Factors Influencing Administrators’ Empowerment and Financial
Management Effectiveness. Procedia - Social and Behavioral Sciences, 176,
466–475.
Dara, S. (2013). Enhancing financial performance with social media: An impression
management perspective. Decision Support Systems, 55, 911–918.
94 RISTI, N.º E5, 07/2016

Urbanization Moderate scale Management of Land and

Grain Production based on Evolutionary Algorithm
Tao Zhai, Yadong Fan
neauzht@126.com
Northeast Agriculture University, Haerbin 150030, China

Pages: 95–105
Abstract: Evolutionary algorithm is an optimization process of simulating

biological evolution, which is a hot research topic in the field of artificial intelligence
in computer science. In this paper, we study the relationship and dynamic effect of
urbanization appropriate scale operation of land and grain production. The result
indicate that; 1.There is no the granger causality relationship between urbanization
and grain production; 2. There is unidirectional granger causality between
urbanization and appropriate scale operation of land, There is a unidirectional
granger causal relationship between appropriate scale operation of land and grain
production; 3. Urbanization in the short term on grain production has a positive
effect, long-term negative impact. Urbanization is the negative influence to the
appropriate scale operation of land; 4.Urbanization is mainly influenced by its
fluctuation; grain production was mainly affected by its fluctuation, but the influence
is weakened gradually, The land moderate scale is mainly affected by Urbanization.
So there is a long-term stable relationship between appropriate scale operations of
land grain production land urbanization.
Keywords: Evolutionary algorithm, Land economy, Urbanization, Grain
production, VAR model
1. Introduction
Land is the foundation of human survival, is the most important agricultural production
material, the protection and utilization of land for stability and improves agricultural
productivity; ensure national food security is of great significance. With the accelerated
process of urbanization in China, constantly improve the level of urbanization, in
this process, land use pattern also constantly changing, land scale management level
continuously improve is an important performance of this change. Although the scale
management of land and food production between whether there is significant positive
related academia is still controversial, however, the 18 big reports of the party clearly
pointed out: to promote industrialization and urbanization of the coordination between
the benign interaction, urbanization and agricultural modernization, and promote the
simultaneous development of industrialization, informatization, urbanization, agricultural
modernization. To speed up the development of modern agriculture, enhance the ability
of comprehensive agricultural production, to ensure the effective supply of national food
security and important agricultural products. Development of various forms of scale
RISTI, N.º E5, 07/2016 95

Urbanization Moderate scale Management of Land and Grain Production based on Evolutionary Algorithm
management, build intensive, professional, organized, and social integration of new

agricultural business entities. Accelerate the construction of urbanization cannot be at the
expense of food production, urbanization to protect food security as the foundation, always
tighten the string of food security(Xu,2011; Pereira, 2015). Heilongjiang Province is China’s
major grain producing area and an important commodity grain base, in-depth study of
Heilongjiang Province to promote urbanization in the process of land scale management
and food production, not only conducive to the promotion of Heilongjiang Province
Rural Economic and social development and for the protection of national food security,
maintenance of national economic and social stability is of important significance. Firstly,
literature on domestic and foreign about urbanization and scale land management and food
production to sort out, to find out the academic circles for the three relations reached a
consensus and existing differences, secondly from theory on urbanization and scale land
management and food production between internal mechanism, and then use statistical
data in Heilongjiang Province of Heilongjiang Province urbanization and scale land
management and food production by empirical analysis, and finally come to the conclusion
put forward corresponding countermeasures and suggestions.
Domestic and foreign experts and scholars have conducted extensive research on the
relationship between urbanization and land scale management. Hong (2005) believes
that the greatest feature of urbanization is the agglomeration effect, the transfer of rural
population is conducive to the land in the hands of farmers moderate concentration,
to achieve the scale of agricultural operations and intensive management. In the long
run, the development of urbanization is conducive to the protection of arable land,
the reduction of agricultural population will help the development of agricultural
scale management and improve the productivity of agriculture. Qu(2011) put forward
urbanization to reduce the agricultural population, realize the scale, modern, intensive
land management. A good, standard land circulation mechanism is the necessary
condition to realize the land management of scale. Huang(1998) pointed out that with
the advance of industrialization, urbanization and rural population transfer and the
rural land contract and management rights gradually circulation and concentration.
With the further transfer of the rural population, the rural circulation to undertake the
management right of farmland area will continue to expand. From the existing literature,
we can see that the relationship of urbanization and land scale operation, scholars draw
a more consistent conclusion, consensus, that the urbanization process accelerated,
urbanization rate continues to improve, conducive to the scale of operation of rural land,
intensive management, contribute to the improvement of the efficiency of agricultural
production, but the conduction of the relationship is the need for certain conditions.
There are many factors that affect grain production, and the research on the influence
of land scale management on grain production has been studied extensively and deeply.
At present, there are two kinds of views, a view that land scale management is able to
overcome the adverse impact on food production of land, fragmentation, land scale
management and grain production has a significant positive relationship; another view
that China’s grain production does not exist significant economies of scale, agricultural
production is constant returns to scale characteristics. With the implementation of
the rural household contract responsibility system, the enthusiasm of the farmers has
been greatly improved, and the grain output has been greatly improved. Household
management has also brought the land use of “fragmentation”, the land fragmentation
96 RISTI, N.º E5, 07/2016

has negative impact on Grain Yield and statistically very significant (Zhang, 2010),
reducing the farmers’ grain production efficiency, reduced grain yield. Subsequently,
the academic circles advocated the adoption of appropriate scale of land management
to improve grain production. In Jiangsu province and Zhejiang Province on the scale
of agricultural operations conducted an experimental investigation, found that large
rice per unit yield higher than the village average of 8%, 3% higher than the wheat.
Chen(2012) using the statistical data of Changshou City in Jiangsu Province, using
production function and partial correlation analysis to draw the conclusion that the
scale of cultivated land management has a significant positive effect on grain yield.
2. Evolutionary Algorithm
Evolutionary algorithm is based on Darwin’s theory of evolution by simulating the
process of biological evolution and mechanism of the problem of self - organization,
adaptive artificial intelligence technology. Biological evolution is realized through
reproduction, variation, competition and selection, and the evolutionary algorithm is
mainly to solve the optimization problem by selecting, recombination and mutation of
these three kinds of operations. Evolutionary computation is a kind of search algorithm
based on natural selection and natural genetic mechanism. As the same with the ordinary
search method, evolutionary computation is a kind of iterative algorithm; the different
is evolutionary computation in the optimal solution search process, usually from a set of
the original problem solution of improvement to another group of good solution, from
Figure 1 – The basic framework of evolutionary

algorithm
RISTI, N.º E5, 07/2016 97

the improved solution of further improvement. And in the evolution of the problem,
when the optimization model of the original problem is established, the solution of
the original problem must be coded. Evolutionary computation in the search process
using structured and random information, so that the most satisfied with the goal of the
decision to obtain the maximum survival possible, it is a probabilistic algorithm.
Generally speaking, evolutionary computation for solving includes following steps:
given a set of initial solutions, the solutions of the performance evaluation; from the
current this set of solutions in choose a certain number of solution as the basis for
iterative solution; again on the operation, iterative solution; if these solutions satisfy
the requirements then stop, otherwise will be obtained by the iterative solution as the
current solution to operation.
Evolutionary computation includes 4 typical methods of genetic algorithm, genetic
programming, evolutionary strategy and evolutionary programming. The first kind
of method is more mature, and has been widely used. The application of evolutionary
strategy and evolutionary programming in scientific research and practical problems is
more and more widely. The main genetic operation of genetic algorithm are selection,
crossover and mutation, and in the evolution rule and strategy, the evolution mechanism
of source on selection and mutation. In terms of fitness, genetic algorithms are used
to select a good parent, and evolutionary rules and evolutionary strategies are used to
select offspring. Genetic algorithm and genetic programming emphasizes the parent
offspring genetic chain, and rules of evolution and evolutionary strategy focuses on
the sub generation behavior itself, immediately as a chain. Evolutionary rules and
evolutionary strategies are generally not used to encode, eliminating the process of
encoding and decoding procedures are more suitable for continuous optimization
Figure 2 – Evolutionary Algorithm
98 RISTI, N.º E5, 07/2016

problems, but it cannot be non-numerical optimization. Evolutionary strategies can be

used to determine the mechanisms that produce the parent generation for reproduction,
and the genetic algorithm and evolutionary rules emphasize the dependence of
individual fitness and probability. In addition, evolutionary rules are abstracted to the
similarity between populations, and the evolutionary strategy is the similarity between
individuals. Evolutionary strategies and evolutionary rules have been applied to many
fields of continuous function optimization, pattern recognition, machine learning,
neural network training, system identification and intelligent control.
3. Research method and variables

3.1. Method selection and variable
VAR model is often used to predict the time series system and analyze the dynamic
impact of random disturbance on the variable system, so as to explain the impact of
various economic shocks on the formation of economic variables. In this study, based
on the test of the stability of the variables, the variables were analyzed by using VAR
model and variance decomposition, to investigate the dynamic relationship between
urbanization, land moderate scale management and grain production.
•• Urbanization level (UR): Urbanization level is an important indicator to
measure the level of urbanization in a country or region. Combined with the
existing research, we found that there are 3 main methods to be used: (1)
population index (2) land index method (3) urbanization index system method.
In the three methods, more and more scholars use population index method,
urban population accounted for accounted for the proportion of the total
resident population and non-agricultural population household population
proportion, the urban population accounted for the proportion of the total
resident population to measure method of level of urbanization has been
accepted by most scholars, and because data are easy to obtain, so this study by
urbanization of the population accounts for the proportion of the total resident
population to represent the level of urbanization.
•• Moderate scale management of land (LS): The moderate scale
management of land is a kind of management way that the operating cost is
reduced gradually by expanding the area of land management. Existing index
of land moderate scale management mainly include worth arable land, labor,
the average cultivated land area and per capita cultivated land area. Based
on the availability of data, this paper selects the rural household average per
person to measure the index of cultivated land area to measure the degree of
moderate scale management of land.
•• Food production (CQ): The index of grain production has two indexes of
total yield and per unit area yield. This paper selects the total grain output in
this paper as a measure of grain production.
3.2. Data sources and descriptive statistics

This study selected in Heilongjiang Province from 1990 to 2011 years of sample data, all data
sources in the calendar year “Heilongjiang statistical yearbook and the variables were log
processing, calculation and analysis were in operation eviews6.0 based. Since the 90’s of last
RISTI, N.º E5, 07/2016 99

century, the urbanization process of Heilongjiang province has been accelerating, and the
urbanization level has been continuously improved. 1990 Heilongjiang province urbanization
rate of 48%, in 2011 the rate of urbanization increased to 56.5%, higher than the national
level of 5.3 percentage points. At the same time, Heilongjiang Province, the average per capita
operating area of cultivated land increased steadily, in 1990 for 7.46 acres, in to reach 10.42
acres in 2005 to reach 10 acres, in 2011 increased to 12.85 acres. In 1990 to 2011 years in
Heilongjiang province’s total grain yield except for individual years slight decline, the overall
trend of rising trend, the grain production in 1990 for 2312.5 million tons, in 2008 exceeded
40 million tons, exceeded 500 million tons in 2010, reached 5012.8 million tons, become the
country the second billions of pounds of production of province super. In 2011, Heilongjiang’s
grain output was more than a grain of grain, for the first time in Henan Province, which
became the first province of grain production in the whole country.
Figure 3 – Trend of the urbanization rate, per capita household management of cultivated land
and grain yield in Heilongjiang province (1990-2011)
4.1. Stability test
Due to the non-stationary time series analysis may produce spurious regression phenomenon.
Therefore, the time series analysis requires the stationarity test and inspection methods are
mainly ADF test and PP test methods. In this paper, the use of more extensive ADF test
method for time series stability test, the specific test results as shown in table 1. As can be
seen from the table 1, the time series LNUR, LNLS and LNCQ in 5% of the significant levels
are non-stationary sequence, its first order difference after the three variables for a smooth
sequence, that is, the first order of the entire sequence, denoted as I (1).
time series type ADF 5% critical value P value conclusion

LNUR (C T 0) -3.4416 -3.7105 0.0790 Not stable
LNLS (C T 0) -0.3210 -3.6584 0.9835 Not stable
LNCQ (C T 0) -1.4616 -3.6450 0.8103 Not stable
△LNUR (0 0 0) -3.0697 -1.9591 0.0040 stable
△LNLS (C T 0) -8.0953 -3.6585 0.0000 stable
△LNCQ (0 0 0) -4.2391 -1.9591 0.0002 stable
Table 1 – Stationary test results of time series
100 RISTI, N.º E5, 07/2016

4.2. Co integration test

The purpose of co integration test is to determine whether there is a long-term equilibrium
relationship between variables. In order to determine the model parameters has more
explanatory power, must be to seek the balance between the lag period and the degree
of freedom, this paper based on the LR test statistic, FPE, AIC information criterion, SC
information criterion and HQ information criterion five indicators of lag order selection.
Results are shown in Table 2. The results show that the selection of the 2 criteria in the late
period of the 4 order delay, the selection of 3 criteria has 1, in order to meet the criteria for
the selection criteria, we consider the model of the optimal lag order of 2 order.
Lag LR FPE AIC SC HQ

0 NA 3.62e-08 -8.6201 -8.4714 -8.5953
1 63.64446 1.37e-09 -11.9161 -11.3196 -11.8151
2 17.57705* 8.92e-10* -12.4335 -11.3896* -12.2568*
3 9.114994 1.07e-09 -12.4989* -11.0077 -12.2465
Table 2 – Selection results of lag order

The characteristics of the root of urbanization, land scale management and food
production of three variables cointegration test. Test results are shown in table 3. From
the cointegration test results, we found that in the absence of cointegration equation
and at least a cointegration equation of the trace statistics at the 5% significance level
rejected the null hypothesis, which shows that there exists a long-term co integration
relationship between LNUR, lnls and LNCQ three variables that between 1990 to 2011
in Heilongjiang Province urbanization and scale land management and food production
exist long-run equilibrium relationship.
integration equation characteristic value Trace statistic 5% critical value P value

None * 0.7349 44.0695 29.7971 0.0006
At most 1 * 0.5750 18.8468 15.4947 0.0150
At most 2 0.1275 2.5919 3.8415 0.1074
Table 3 – J-J cointegration test results
4.3. Grainger causality test

The co integration test shows just there is a long-term equilibrium relationship between
the urbanization and scale land management and food production, but the relationship is
a causal relationship, still need further do Granger causality test. Test results are shown
in table 4. The analysis of Grainger causality test results the following conclusions can be
drawn: first, significant at the 10% level, the scale management of land grain production in
LNLS is Grainger LNCQ, Grainger LNCQ is not a grain production scale land management
LNLS, there is a one-way causal relationship between land scale management and food
production LNLS LNCQ, which is the early information of land scale management LNLS
will affect the current grain production of LNCQ; second, there is no causal relationship
between Grainger and the grain production in town; third, at 5% level, urbanization causes
RISTI, N.º E5, 07/2016 101

the land scale management LNUR is Grainger LNLS, Grainger LNLS is not the scale
management of land urbanization LNUR, with a one-way causal relationship between
urbanization LNUR and LNLS scale land management.
hypothesis sample F statistics P value conclusion

LNLS does not Granger Cause LNCQ 20 3.0037 0.0800 refuse
LNCQ does not Granger Cause LNLS 1.1799 0.3343 Not refuse
LNUR does not Granger Cause LNCQ 20 0.3701 0.6968 Not refuse
LNCQ does not Granger Cause LNUR 2.0932 0.1578 Not refuse
LNUR does not Granger Cause LNLS 20 4.9153 0.0228 refuse
LNLS does not Granger Cause LNUR 2.0801 0.1595 Not refuse
Table 4 – Granger causality test results
4.4. Impulse response analysis

The impulse response is analyzed when an error term is changed, or when the model is
subjected to a certain impact on the system dynamics. Figure 4 is the impulse response
graphs on Heilongjiang Province urbanization and scale land management and food
production, the horizontal axis of the graph represents the is impact lag periods and the
vertical axis represents the impulse response variables, the dotted line is standard deviation
band, the solid line shows the impulse response function. It shows urbanization of their
own a standard difference of positive innovation, has the positive effect in the first period,
followed by a substantial decline, in the fifth period before have the positive effects, but from
the beginning of the fifth to the tenth stage has been negative impact. The urbanization on
the land scale management positive new interest and a standard deviation, the influence
of the first phase as 0, leveled off after reached the maximum in the third period, the
overall positive influence. This shows that there is a long-term close relationship between
land scale management and urbanization, land scale management has a positive role in
promoting urbanization. The urbanization of grain production of a standard deviation
of the positive new information, in the first phase of the impact as 0, in the third period
reached the maximum after the decline, but the overall positive impact. This shows that
there is a long-term close relationship between urbanization and food production, in the
short term urbanization had a strong response to food production, from the long run this
response is gradually weakened, but still has a stable and positive role in promoting.
Figure 4 – Response function: LNUR to LNUR, LNUR to LNLS,LNUR to LNCQ
102 RISTI, N.º E5, 07/2016

Figure 5 shows that the scale of the land management of a standard deviation of the positive
new information, in the first phase has a negative impact, followed by the negative impact
of each period, and the negative impact of the gradual increase. Land scale management
of their own a standard difference of positive innovation. In the first period has positive
effect, began to exhibit wavy effect from the first stage, even down to a trough, the crest
of odd period rise, the trough value and peak value is basically the same, on the whole is
now positive effect. Land scale management on grain production in a standard deviation,
effects in the first period of 0, second period has a small positive effect, third stage and
after each period appear wavy effect, but are slightly negative effect.
Figure 5 – Response function: LNLS to LNUR, LNLS to LNLS, LNLS to LNCQ

Figure 6 shows that grain production of town of a standard positive innovation, has
significantly positive effect in the first period, then declined gradually, starts from the
fourth to the tenth period were negative effect, and the negative effect gradually increases.
The grain production of land scale operation of a standard forward new information, in
the first phase of the impact, in the second period to the lowest after a slight rise to
tenth, a positive impact on the overall. The grain production of a standard deviation of
their own, in the first phase has a significant impact, and then decreased to sixth for 0,
followed by a negative impact on the tenth.
Figure 6 – Response function: LNCQ to LNUR, LNCQ to LNLS, LNCQ to LNCQ
5. Conclusions
Based on the statistical data of Heilongjiang Province, this paper studies the relationship
between urbanization and moderate scale management of land and grain production
in province. Mainly obtains the following conclusions: (1) despite the urbanization
RISTI, N.º E5, 07/2016 103

rate, moderate large-scale management of land and grain production of non-stationary

time series, but after the first order difference tends to be stable, using five information
criterion to determine the optimal lag period 2 order. (2) Co-integration test shows that
there is a long-term stable relationship between urbanization, land scale management
and grain production. (3) Granger causality test results show that there is no Granger
causality between urbanization and food production; there is a one-way causal
relationship between urbanization and land scale operation, at 5% significant level,
urbanization is Granger cause of appropriate scale of operation of land; land moderate
scale management and food production is a one-way causal relationship also exists
between, at the 10% level of significance, the appropriate scale operation of land is the
Granger cause of grain production. (4) From the analysis of impulse response function
curve, we can know that there is a positive effect of urbanization on grain production in
the short term. Urbanization has a negative impact on the moderate scale management
of the land; the moderate scale management of land has a strong positive impact. (5) The
variance decomposition shows, urbanization is mainly affected by its volatility effect.
This effect remained at more than 58%; grain yield mainly by its fluctuation effect, but
the effect gradually weakened, in the four urbanization has a great influence on the grain
yield; appropriate scale land is mainly affected by urbanization, overall, three effects of
moderate scale management of land relatively stable.
This paper reveals that the urbanization, moderate scale management of land and food
production is not isolated, in the future development strategy to correctly handle the
relationship between the three. Based on the relationship between urbanization and
grain production, we should steadily push forward the urbanization and ensure the
national food security. Urbanization can promote the grain production in the short
term, however, the promotion effect is gradually weakened, and the urbanization has a
negative effect on grain production in the medium and long term. The reason may be in
the short term the rural surplus labor transfer to urban and bring the improvement of
the efficiency of grain production, with much of the labor force flow to urbanization, the
improvement of grain yield influenced and restricted by the shortage of labor. Based on
the positive pulling effect of moderate scale management of land and grain production,
and actively promote the appropriate scale of land management, both to improve the level
of urbanization, there can be to promote food production. Establishment and improve
the land transfer system, cultivate the market of circulation of rural land contracted
management right, land supply and demand information smooth and symmetrical,
encourage landowners to the grower and business experts, implementation of grain
production scale, intensive management.
Acknowledgments
The work of this paper is supported by Philosophy and Social Science Foundation of
Heilongjiang Province (13C012); Applied technology and research and development
program of Heilongjiang provincial science and Technology Department (GZ13D103).
104 RISTI, N.º E5, 07/2016

References
Chen, X. (2012). The “three rural” issues in the process of urbanization in China. Journal
of National School of Administration, (6), 4–11.
Hong, B., Chen, A. (2005). The urbanization and the “three rural” solve the problem.
Finance research, (4), 24–29.
Huang, Z., Chen, X. (1998). Grain farmers scale efficiency: Empirical Analysis and some
conclusions. Problem of agricultural economy, (11), 2–7.
Qu, H., Xia, L. (2011). The relationship between the development of China’s tourism
industry and economic growth empirical research. Finance & Economics, (8),
106–112.
Xu, Q., Yin, R. (2011). Economic Research on scale economy, scale returns and
agricultural moderate scale management. Economic research, (3), 59–71.
Zhang, Z. (2010). An empirical study on the relationship between farmers’ land
management scale and grain production efficiency. China land science, (8), 52–58.
RISTI, N.º E5, 07/2016 105

E-course of College Badminton Curriculum Based on

Multimedia Platform: an Experiential Teaching Method
Hong Zhang
3570313@qq.com
PE School, NanJing XiaoZhuang University, Nanjing 211171, China

Pages: 106–114
Abstract: Multimedia technology has the powerful influence in the field of modern
sports education, a new education method as experience teaching method will
better promote the teaching effect. By using empirical analysis, the result shows that
experience teaching method will increase the students’ interest in learning. More
than 50% students increase interest in badminton class through the experience,
also the experience teaching method will promote students’ physical quality and
basic badminton skills. Therefore, the experiential teaching will stimulate the
initiative of students learning; and will effectively improve the teaching quality of
badminton courses. On this basis, we put forward relevant proposals.
Keywords: Multimedia education, experiential method, badminton curriculum,
online questionnaire
1. Introduction
With the continuous progress of China’s education, deepening education reform and
development of physical education, scientific research has been fruitful. However the
common orientation of modern education and teaching practice is to promote students
all-round and harmonious development, improve the students’ comprehensive quality,
and achieve the teaching goal, the nation each big institutions of higher learning must
be continuous innovation, accumulate experience, summed up a set of applicable in
nowadays teaching methods to meet the demand the stages of contemporary college
students, of course, also give us badminton teaching brought new opportunities and
challenges(Zhang,2012;Dai, 2012). Experience type teaching method is applied to the
badminton teaching process, can make class teaching more lively, and for students to
create a relaxed and harmonious atmosphere in the classroom. Experience of classroom
teaching is by a series of conducive to the positive transfer of technology of experiential
sports to guide the students to actively participate in the beginning of experience, with
strong entertainment atmosphere in the classroom began learning technical movements
in the, such not only can fully mobilize the enthusiasm of the students enthusiasm in
learning this lesson, can further improve the interest of learning the sport of badminton,
thus contributing to the future study of the teaching content(Hu, 2012; Kim,2014;
Crespo,2015). The rise of badminton sport in China is the embodiment of the people
sports concept and sports function. The sport to further enrich the content of physical
106 RISTI, N.º E5, 07/2016

exercise, have also gradually become a hot item in the domestic and international mass
sports and the modern sports fitness campaign of new features. This study is of mass
badminton teaching reform and innovation, have important significance to break away
from the universities and colleges in the traditional badminton skills teaching ideology
(Sigala, 2012; Jian-hua,2012; Krstev, 2014). More and more the phenomenon of the
current sports public elective course in Colleges and universities to carry out and not
particularly optimistic, many students think the sports class is boring, the enthusiasm is
not high, are not keen interest in learning; students appear withdrawn, alienation and
other psychological problems, students analyse and solve the problem of capacity needs
to be improved. So in order to adapt to the rapid development of social requirements for
talent, we need to a take the student as the main body, pay attention to the teacher-student
interaction, lively, interesting and strong, for the non-sports professional students this
group of character and thinking characteristics of the teaching methods. It is more likely
to help students to change the idea, in order to obtain a more solid knowledge.
With the popularity of badminton sport in recent years, in the sports elective course
in Colleges and universities, badminton course has attracted a lot of the majority
of students love, teachers how to better carry out badminton class, the use of more
reasonable and effective teaching methods have become the most important badminton
elective. In the process of teaching teachers cannot blindly focus on how the experience
knowledge transfer to the student’s mind, and to let the students learn to ask for, from
their own personal experience summed up the experience of truth and to be integrated,
to the teachers to obtain knowledge and ultimately more reasonable effective will be on
the understanding. The experience teaching method is based on the traditional teaching
methods, students acquire knowledge through experience, and meaning and emotion
in learning constantly reflect experience, life experience and ability character (Huang,
2013). Therefore, attempt by the ordinary university badminton option class in the use
of experiential teaching methods for teaching and research, to a certain extent in order
to further improve the quality of teaching in the teaching of badminton. In order to
promote the sports undertakings flourish, and thus stimulate student’s sports interest
in learning, the establishment of lifelong sports consciousness, this study tries to from
thebadminton elective class exploring experience type teaching method application and
implementation. With its from teaching and formulate relevant syllabus and all kinds of
teaching documents in order to provide reference for the major colleges and universities,
and have used this method of colleges and universities in improve the university
badminton teaching quality, improve teaching file to provide some of the material.
2. Multimedia teaching platform

Badminton is a general physical education course in Colleges and universities. The basic
technology includes the steps of lower limb, the technique of upper limb and the serving
method. Because badminton movement is delicate and complex, students should
grasp and improve the technical and tactical level of the project. How to innovate the
badminton teaching method, give full play to the leading role of teachers and students,
through access to relevant information to show that the application of multimedia
assisted instruction is one of the main directions of sports teaching reform. Multimedia
technology as a supplementary means of badminton teaching has many advantages,
RISTI, N.º E5, 07/2016 107

E-course of College Badminton Curriculum Based on Multimedia Platform: an Experiential Teaching Method
but badminton is a practical course, it is impossible to replace the traditional teaching

methods. Students through the multimedia courseware to watch, for the formation of the
concept of action, the formation of the action representation, correct the wrong action
to get a positive effect. But the badminton as the ultimate goal of curriculum practice is
to master the technical movements, achieved through traditional teaching. Therefore,
multimedia technology just auxiliary means in the traditional teaching process, it
cannot replace the traditional teaching method. In order to receive good teaching effect,
multimedia technology should be combined with traditional teaching method.
Figure 1 – Badminton simulation in 3D environment
Figure 2 – The basic framework of badminton teaching

In badminton teaching, teachers of traditional explanation and demonstration has some
limitations, especially the contradiction between the demonstration action time and
space, such as off air smash and demonstration may not stay in the air for students
to see the details. And multimedia technology breaks through the space and time
advantage, to freeze the details of its action, labeling, animated illustrations, students to
provide intuitive, image of a variety of demonstration for students to understand clearly
every action link, promote students to establish correct action concept and clear action
108 RISTI, N.º E5, 07/2016

representation. Therefore, the use of multimedia assisted teaching than single operation
with traditional teaching in the explanation and demonstration of make it easier for
students to understand and master the technical movements and make the teaching
more rational and scientific.
Figure 3 – The badminton teaching
3. Experience teaching method of badminton

3.1. Design of experiential teaching method
Experiencing type teaching method is to practice thinking development of students as
the goal point, to the effective solution of practical problem as the basic support, so that
teaching activities become the organic unification of teacher’s guidance and students’
learning activities. In the whole process of teaching experience teaching penetration in
the whole teaching process, develops take the student as the center, according to the
students’ interests and individual differences of autonomous learning and teachers’
guidance, combining, and overall development of the students’ physical and mental
health and improve the teaching efficiency of the teachers.
1. Beginning of the class: teachers according to the students in the
experimental group, in accordance with the teaching task, the interest hobby,
exercise capacity, physical quality and teaching requirements put the students
in different groups, when the students in their respective groups to experience
type badminton, to them were classified teaching guidance, establish action
Essentials, to establish the standard of assessment. Before the students to study
and practice to provide background, set up the teaching problems, inspire
students before the experiential exercises of independent thinking, and promote
the understanding of students’ psychology of badminton skills and knowledge,
stimulate students’ interest in and desire for knowledge, and content to students
clear the students, students in each group were to experience practice.
2. In the course: according to the instructions of teachers teaching, students
learn from each other, learn from each other in their respective groups. Aiming
RISTI, N.º E5, 07/2016 109

at the error action of the students in practice and the special problems, such as
hit the lofty ball hit by, twist is not sufficient, timely feedback to teachers, find
out the reasons, suit the remedy to the case, and give the error correction. For a
variety of issues and questions raised by the students, to give specific guidance,
prescribing exercise designed to facilitate students’ learning and improvement
and is conducive to arouse students interest and experience a variety of ways
and means of practice. According to the guidance of teachers students exercises
to consolidate and improve, achieve a multiplier effect.
3. After class: experience type teaching method in addition to the application
in the teaching process, in class also has wide application, by teachers to design
the experience type teaching program, students can in spare time using class a
more profound impression on experiential teaching method of repeated practice
to imitate, so as to deepen the technical movements in the master. Teachers
can appropriate arrangements for after-school exercises, through experiential
exercises gradually transition to complete technical exercises, effectively
consolidate and improve the degree of mastering the technical movement.
3.2. Effect of experiential teaching on learning interest

Students interested in the badminton class degree shows, through to the experimental
group and control group students questionnaire investigation that, experimental group
have to badminton class nearly 50% of the students are very interested in nearly 25%
of the students of Badminton Class expressed interest. But the control group was only
about 32% of the students are very interested in badminton, interested in badminton
class students accounted for nearly 27%. The experimental group and the control group
of badminton course said very interested or interested in the proportion of students
is 16%. The experimental group of badminton class is not interested or not interested
students is almost 0, but the control group students of badminton class is not interested
or not interested in the proportion of more than 10% students. Through the experience
type teaching method of the teaching experiment, the experimental group and the
control group in the level of interest of Badminton Class showed significant difference.
The students in the experimental group through a semester of experiential learning can
have a great interest in badminton class. The results show that the experiential teaching
method can improve students’ learning interest in badminton, which has promoted the
popularity and development of badminton in a certain extent. Today is the badminton
movement the most popular time, good badminton skills can not only achieve physical
fitness, more able to meet the needs of contemporary people to pursue fashion? At
the same time, in contemporary people gradually pay attention to physical exercise of
society, palm grip a sport skills into the communication between people of the most
effective way. Therefore, in the sports option class, badminton class into the one of
the majority of students love project. However, teachers should want to in the original
foundation to further improve the attitude of students towards badminton, requires
teachers with good teaching methods to guide, experiential teaching method fully
embodies the teachers’ effective guide and promote the students’ interest in training. So
110 RISTI, N.º E5, 07/2016

the experience type teaching method plays an important role in the popularization and
development of badminton sport.
4.1. Physical fitness test
Physical education teachers in the school sports science experts, curriculum theory
expert, and ordinary institutions of higher learning and Dean of faculty, colleges and
universities are a large number of interviews, access to a large number of a wealth of
information about experiential teaching method in badminton as a selective course in
application. In particular, in-depth interviews of College P.E. teachers, understanding
of the experiment in the presence of difficulties and problems, for the application
of experiential teaching methods to provide the objective and real data, make the
experience type teaching method in badminton as a selective course in the application
is more practical and fit. According to the specific research direction, through access to
a lot of literature and combined with my own experience design two copies of the first
draft of the questionnaire, followed by visits to relevant experts, analysis and review
of many experts, I also according to the opinions and suggestions of the experts, the
questionnaire were finishing, modify and perfect, eventually forming can effectively
reflect the badminton to students learning interest and teaching and the special quality
index of the formal questionnaire. In order to determine the experimental group and
the control group homogeneity, in order to ensure the validity of this experiment has
high, before the experiment of experimental group and control group were pre-test, test
content includes the student’s basic physical fitness, badminton skills test, badminton
first began to learn learning motivation survey. In addition, our results on students’
physical test summary table as follow:
designated
Standing long 800m
group solid ball X±S shuttle run
jump X±S X±S
X±S
Experimental classes 177.45±7.81 5.62±0.64 21.25±1.05 3.49±0.19
control group 182.66±7.88 5.45±0.65 21.42±1.02 3.53±0.21
T 2.437 3.455 2.258 2.057
P >0.05 >0.05 >0.05 >0.05
Table 1 – Analysis of physical fitness test
Before the experiment, were experimental group of NanJing XiaoZhuang University and
control group students the forehand the lofty ball, forehand hitting the lofty goals and
forehand network before rubbing ball test. Forehand made the evaluation criteria of the
lofty goals that serve to singles and doubles service line between effective; forehand hit
RISTI, N.º E5, 07/2016 111

the lofty goal of evaluation criteria: to play to the singles and doubles service line for
effective; forehand before rubbing ball evaluation standard for: placement of the ball is
the distance the net vertical line one meter or less for. Required tests shall be carried out
technical checks by experienced badminton teachers; specific operation is completed by
him. The experimental group and the control group badminton single technology test
results as table 2:
Index Experience group Control group T P

Lofty ball 4.18±0.51 4.26±0.81 0.49 >0.05
Hit lofty goals 5.75±0.71 6.12±0.83 1.05 >0.05
Chopping 5.18±0.63 5.73±0.82 1.11 >0.05
Table 2 – Badminton single technology test results
4.2. Badminton technology evaluation

An essential feature of the experience teaching method is to stimulate the students’
emotion, to promote students’ cognitive activities. And teaching demonstration law only
the objects, teaching aids presented to students or teachers simply do demonstration
experiments, although there are intuitive, but only object can only lead to students
cold intelligence operations, and can’t play of emotion. A comparative analysis on the
influence of the experience teaching method and the traditional teaching method of
the teaching quality of badminton elective single technology, test results and analysis
are shown in table 3.
Index Experience group Control group T P

Lofty ball 7.35±1.52 5.17±1.09 0.0022 <0.01
Hit lofty goals 7.25±1.42 6.35±1.43 0.0029 <0.01
Chopping 7.85±1.65 6.47±1.03 0.0010 <0.01
Table 3 – Badminton single technology after the experiment
In the badminton elective course teaching, integrated test evaluation can effectively
reflect student learning effects. In the semester curriculum, teachers should properly
organized teaching competition, this will enable students to be learned in the classroom
of the technical action of stage of consolidation and further improve, at the same time,
in various stages of teaching the game to make the students timely found the existence
of their own lack of, and feedback to the discussion of the shortcomings and deficiencies
through the communication between students and teachers, and ultimately to be
corrected. The following is a comprehensive test of the university badminton elective
course of the result of the specific evaluation index, as table 4.
112 RISTI, N.º E5, 07/2016

Group Number X±S T P

Experience group 33 82.04±7.66 3.172 <0.05
Control group 33 77.12±8.15
Table 4 – Comprehensive test results

In the experience type teaching, teachers through the organization of arrangement
reasonable teaching the game to enable students to understand a certain stage of their
own, teaching competition here is not only refers to the game of badminton’s teaching,
also includes the experience type teaching competition, teachers in order to promote
the students’ understanding of the teaching content, before class design experience.
However, in order to further enhance the students’ interest and understanding of the
teaching content, teachers can will experience a type scheme is presented in the form
of game teaching. The student experience at the same time, and can feel the joy of
success. The average integrated test can effectively reflect the students throughout the
semester learning effect, and through the concrete test results can be timely feedback
to the students their own shortcomings and deficiencies exist. However, through the
experimental study of experiential teaching methods show that experiential teaching
method can effectively improve the level of comprehensive testing of students. Students’
self-evaluation table as shown in table 5:
Group Fit High Low Total number

Experience group 25 4 4 33
Control group 14 9 10 33
Table 5 – Students’ evaluation of classroom teaching content
5. Conclusions
In the teaching process, learning attitude of teachers’ teaching methods and teaching
art students have a more important influence. In badminton teaching process, students’
learning attitude directly affects the teachers’ teaching and students’ learning process: a
good learning attitude can help to teachers in teaching give full play to its advantages, and
with a more positive attitude to create a more perfect teaching plan. At the same time,
students have a good learning attitude, will directly promote the students’ learning effect
is good or bad; therefore, this study through the student self-evaluation of badminton
teaching content and to the badminton teaching classroom group.
Evaluate the organization of learning attitude of students, so as to achieve the purpose
of testing the experience type teaching method. Each student due to many reasons
of their growth environment, cognitive level, and interest, hobbies and so lead on all
aspects of the teaching factors have different understanding, will eventually manifested
in the understanding of classroom learning. Therefore, in the experience of teaching
process, the teacher will give full consideration to provide independent thinking, and
create opportunities for students, and let them experience the joy of success in the
RISTI, N.º E5, 07/2016 113

performance of the process. At the same time, students in experiential education of

the learning process, can further improve their own shortcomings and deficiencies,
teachers will also be arranging different experiential scheme according to the students
of their own characteristics, fully mobilize all the students enthusiasm for learning, and
improve their cognitive level and theoretical level and practical level. In the experiential
teaching process, teachers will use experience of sports to promote students’ learning
the technical movement, and then follow the sport technique learning rules, so that
the technology learning can proceed in an orderly way, interlocking and step by step.
However in the teaching, teachers pay more attention to the students of learning
content of active reflection, discuss and share. Therefore, the experience type teaching
method not only in a certain extent is conducive to students’ mastery of technology,
and more conducive to the students to analyze and solve problems ability training. The
requirements of the teaching program of teachers in the teaching process to give full play
to students’ subjectivity, respect students’ individual differences.
References
Crespo, P., Santos, V. (2015). Construction of Integrated Business Management Systems
for Micro and Small Enterprises. RISTI-Revista Ibérica de Sistemas e Tecnologias
de Informação, (15), 35–49.
Dai, W., Fan, L. (2012). Discussion about the Pros and Cons and Recommendations for
Multimedia Teaching in Local Vocational Schools. Physics Procedia, 33, 1144–1148.
Hu, M., Xu, S. (2012). Research of Multimedia Teaching on Principles of Management.
IERI Procedia, 2, 666–670.
Huang, Y., Backman, S. (2013). Experiencing student learning and tourism training in
a 3D virtual world: An exploratory study. Journal of Hospitality, Leisure, Sport &
Tourism Education, 13, 190–201.
Jian-hua, S., Hong, L. (2012). Explore the Effective Use of Multimedia Technology in
College Physics Teaching. Energy Procedia, 17, 1897–1900.
Kim, A., Davies, J. (2014). A teacher’s perspective on student centred learning: Towards
the development of best practice in an undergraduate tourism course. Journal of
Hospitality, Leisure, Sport & Tourism Education, 25, 6–14.
Krstev, C., Trtovac, A. (2014). Teaching Multimedia Documents to LIS Students. The
Journal of Academic Librarianship, 40, 152–162.
Sigala, M. (2012). Investigating the role and impact of geovisualisation and
geocollaborative portals on collaborative e-learning in tourism education. Journal
of Hospitality, Leisure, Sport & Tourism Education, 11, 50–66.
Zhang, C., Chen, X. (2012). Use of Multimedia in Gross Infective Pathogen Experimental
Teaching. Procedia Engineering, 37, 64–67.
114 RISTI, N.º E5, 07/2016

Research on Badminton Footwork Training

and Teaching Effect based on Computer Aided
Information Processing
Chao Chen
chenchao10126@163.com
PE School, NanJing XiaoZhuang University, Nanjing 211171, China

Pages: 115–124
Abstract: Multimedia spots information processing system can be more

rapid processing to all kinds of processing, search, storage and transmission. In
addition, the sports information processing system has a special function, which is
capable of dealing with multi-source sports information. In this paper, the author
analyzes the badminton footwork training and teaching effect based on computer
aided information processing, the result shows that after the footwork training,
the technical level and the physical quality of the students have been improved
obviously. At the same time, the footwork training method can enhance students’
interest in badminton sport; also will improve students’ heart and lung function.
Keywords: Information Processing, badminton training, teaching quality,
technical level
1. Introduction
With the development of computer technology, multimedia computer technology has
been widely used in the field of physical education. The application of this representative
is sports information processing system, which using the technology have multimedia
technology, audio & video technology and communication technology and the use of
this system also greatly enhances the sports information processing, transmission,
retrieval and storage capacity. Badminton sports as our traditional advantages of the
project, in the competitive badminton training level in the world’s leading level (Sigala,
2012; Jian-hua, 2012). With the increasing level of China’s competitive badminton,
led to the development of domestic colleges and universities badminton, badminton
sport is also more and more people. Nowadays, the badminton sport in Colleges and
universities is more vigorous development, swept the country’s major colleges and
universities, badminton sport is one of the most enthusiastic about the topic among
college students. Now the development of badminton sport, mainly with fast speed and
accuracy in placement, stability control, smash ruthless, emphasize the ability of the
technology of comprehensive and surprise attack. This, combined with the demand of
Badminton Footwork, to complete the pedal step, jump ball, diving and so on each kind
of high difficulty action(Zhang, 2012; Hu,2012; Kim,2014). If you do not have good and
RISTI, N.º E5, 07/2016 115

Research on Badminton Footwork Training and Teaching Effect based on Computer Aided Information Processing
stable technical footwork, master good technique to demonstrate the use, in court only
fast moving in place, you can grab highest point, rational hitting point of threatening
to hit the ball, so step technique of badminton sport is very important. It is the premise
and quality assurance of badminton sport and it is also a bottleneck in the process of
technical and tactical improvement for beginners. And the wrong move gait is very
easy to cause the sports injury, hindered the badminton technology enhancement, and
has hindered the badminton movement development (Dai, 2012; Crespo, 2015). And
through the research of this topic will help to improve the ability to move the badminton
step technology, and enhance the level of competition. In preventing sports injury
caused by wrong action at the same time, improve the physical quality of the students;
let students further understand the badminton sport, their own lack of cognition, so as
to rapidly improve badminton athletics level.
Badminton technique is the basic and the most important part of badminton. Most of
the badminton movement of college students will be excessive concentration of energy in
the badminton strokes of the practice, and a disregard for Badminton Footwork practice
(Huang, 2013; Krstev, 2014). It formed in badminton, because step, hitting action will
deformation, directly affects the various batting technique. And in Badminton Footwork
practice, a student of technical footwork movement standardization degree is not very
high. This will affect the rapid development of badminton sport, so in the badminton
training increased Badminton Footwork Training and step standard. According to the
teaching practice and observation, most of the college students have different kinds of
problems in the process of the process. In this way, it hinders the improvement of College
Students’ badminton level. Therefore, through the Capital Institute of physical education
students of Badminton Footwork teaching practice of experimental study and analysis,
thus finding and improve step technology, help college badminton sport level of ascension.
2. Computer information processing

2.1. Sports information
Multimedia computer sports information processing system is the application
technology of multimedia technology and computer, audio-visual and communication
integrated arising from an information processing system capable of faster, better sports
information to all kinds of processing, search, storage and transmission. In addition, the
sports information processing system has a special function, which is capable of dealing
with multi-source sports information. This function can ensure that the system can
effectively deal with multi-source sports information. In addition, multimedia computer
sports information processing system in the sports information is recorded using the
medium is different, because the record different kinds of information that may be
used for the medium is different, for example, is the sports video information need
to use camera recorded on film, and then use professional equipment on the show or
transmission; sound sports information records will need to use the device for recording
the recorded on the cassette; and text, data and other types of sports information can
record the above the computer disk. In the sports information storage, you can use
multimedia equipment and computer software and according to the specific data format
will stored in the computer, according to the different needs of sports information
decomposition or comprehensive treatment.
116 RISTI, N.º E5, 07/2016

Figure 1 – Data collection and processing
2.1. Multimedia computer

Multimedia computer sports information processing system is more advanced multimedia
technology, which makes sports information processing system can effectively spread on
a wide variety of comprehensive analysis and processing and will sound, text and image
data together, and stored in the database system of sports information broadcast on
the screen, or is the use of the network to transmit them. In addition, the system also
has interactive features. This feature is refers to the sports information disseminators
and recipients between real-time information exchange, and exchange process is often
people use system programming and dialogue, and through active control system to
carry out the work, in order to analyze, research has been accepted and storage of sports
information, make sure to really play the role of sports information.
Sports information mainly contains images, text, audio and video and other kinds of
information, it represents the amount of data is very large, so the difficulty of its processing
is also larger. For example: on a resolution of 640 * 480 24 bit color images were stored,
the amount of storage used is 1m bytes and amount of data is change with the changes of
resolution of, when the resolution of the image becomes higher, the amount of data will
become more, and under normal circumstances, a 1 second video picture contains a total
of 30 (NTSC video standards) image. Multimedia technology in dealing with these data,
a large amount of sports information, usually use two kinds of methods: first, the use
of large capacity memory, such as CD, application of gigabit optical disc is the material
guarantee for the multimedia storing large amounts of data of sports information.
Second, the use of data compression technology to reduce the amount of data sports
information. Multimedia computer sports information processing system in the
compression technology of image compression can use broadband network for dynamic
video remote transmission, in order to make the sports information transmission and
access to more quickly, sports information processing system for remote transmission of
information, such as the following figure 2.
RISTI, N.º E5, 07/2016 117

Figure 2 – information processing system

A number of technologies of multimedia computer sports information processing
system can be developed to dig out more and better software system. Such as the use of
application system design platform to design a set of exercise training and consulting
system, after advancing with the times of training theory knowledge, training mode,
movement method and recovering methods information is added in the system, the
sports information processing system of sports training and consulting system will be
able to use this knowledge to solve various problems in the process of training, and
provide reasonable suggestions and a full range of services for sports training, so that
exercise training can become more scientific.
3. Badminton teaching method

3.1. Footwork exercises
Fixed footwork training method is a practice method of Badminton Footwork teaching
and training, it is refers to two or more than two basic technology, through a combination
of certain fixed line into practice method in badminton pitches exercises. It can quickly
let students master single step and relatively complex binding steps, but also let the
students to participate in the route design, lets the student feel footwork exercises will
not be so dull, can effectively improve the students’ interest. Interest for students, for
their study and work is of great significance, it can motivate students to seek knowledge
desire, so as to achieve the autonomy of students, so as to improve the efficiency of
students’ learning. College is the life energy and stamina most exuberant one age period,
and sports class specialized student due to the profession itself, in the energy and strength
of this period to than other students more exuberant, due to the relatively strong in this
age, physical quality, physiological development more perfect. The function is becoming
more and more stable; their physical quality is in a certain peak. And with the growth
of the age of the students, the students are growing their bone cells, more and thicker
bones, bone development is basically completed. At this point, the use of the fixed line
118 RISTI, N.º E5, 07/2016

method of continuous practice can be appropriate to increase the exercise time and the
intensity of confrontation, to promote student bone thickening and bone thickening.
Similarly, the age of the students body muscle composition is also changing, the water
gradually reduced, the protein gradually increased, the quality of the muscles continue
to improve, and gradually developed, muscle strength continuously strengthened. At
this time the student’s own circulatory system basically reached the level of adult, heart
muscle is relatively developed, the heart rate gradually slowed down, the blood pressure
is basically in a stable state. Therefore, in the use of fixed line exercise method, it can be
appropriate to increase the amount of exercise and exercise load, so as to improve the
level of system function in all aspects of the body.
Figure 3 – Badminton training

Fixed footwork training method for students in accordance with the design of the design
of a good route to repeat the practice, so that it generates the appropriate conditions for
reflection, is conducive to better grasp of the students and strengthens the badminton.
Through for students of the load intensity, the student body has a relative degree of
adaptability, is conducive to development and to improve the physical quality of the
students in all aspects; fixed line training method let students continued the running
practice, enhance the speed endurance quality of students, to complete the exquisite
technical movements in a relatively large load strength; fixed line practice make students
cycle step running exercises can inspire students to practice the positive emotions,
enhance high difficulty action of normative and blend into the skill and the special quality
and various body metabolic system practice exercises. Fixed line exercise method is also
followed by the simple to complex, from fixed to semi fixed line exercises in line with
the general objective of the law of exercise training and the basic principles of exercise
training. In addition, in the badminton to learn and practice, specially designed some
special fixed line for sexual practice, help students to have mastered the basic technology
effectively combined use, while the frontcourt and backcourt, offensive and defensive,
straight lines and oblique lines such as technology and consolidate and improve the
students’ level of skill and tactics and combat capability.
RISTI, N.º E5, 07/2016 119

Figure 4 – Badminton footwork
3.2. Basic principle

•• Systematic training principles: the principle of systematic training is a
necessary condition for the continued practice has ideal training effect and achieve
expected effects of exercise, the body’s biological adaptability training load must
be through the human body a little bit implementation. Suitable for the human
body to adapt to the rules of the training exercises, athletes in the biological aspects
of the changes in the benefits of improving, it takes a long time to train and train. It
can be used for a single action exercise, but also can be used for multiple movement
exercises, pay attention to gradually increase the intensity of the exercise of the
exercise, improve the integrity of the quality of the exercise.
•• Appropriate load principle: exercise is composed by the volume and
intensity loading, because training plan, in order to arrange the load is different
and in line with should be according to the different stages of training, will exhibit
different characteristics. The effect of the exercise is to meet the realization of the
individual by applying the appropriate exercise load to the individual. And the
exercise load is composed of load intensity and content of, due to the different
training program, thereby to arrange the load is also different, motion load and
load measuring, according to different training stages and will exhibit different
characteristics. The individual students in practice to bear a certain amount of
load intensity and load, and that the body is bound to produce a corresponding
change in the body, resulting in a certain amount of exercise effect.
•• Theory and practice: in combination with the actual implementing quality-
oriented education and the implementation of education policy must be to do
two, is their own development needs of sports teaching, is the establishment of
student individual right of sports scientific values of real need. In badminton
teaching practice, according to the actual situation of students and the actual
situation of social life, includes two aspects. Students in learning and mastering
the movement of technical knowledge, but also to understand and accept the
sports culture, deepen the understanding and understanding of sports culture.
In the badminton teaching practice, should give full play to the theoretical
knowledge in the technical skills of the important guiding role in learning and
120 RISTI, N.º E5, 07/2016

mastering technical skills at the same time in theoretical knowledge to deepen

understanding and learning.
4.1. Experimental design
Before the experiment on two groups of students to the body shape, physical fitness,
and footwork speed test, if the experimental group and the control group students
in the above aspects, no significant difference, the experimental object is reasonable
and effective, which can carry out the experiment. Ensure the teaching content of
the two groups, the number of teaching needs to be consistent, and the timing
of the time, the consistency of the venue, equipment. In the same teaching plans
and contents, based on the basis of assessment criteria, sports venues and other
conditions, with the traditional step teaching method (led by the teacher) controls
for experimental teaching, the experimental group focuses on fixed line practice
teaching step. After eight weeks, the experimental group and the control group
were given the same physical quality and gait speed test, and the data obtained
were statistically analyzed.
project experience group control group T value P value

100 meters 12.15±0.13 12.16±0.15 -0.142 0.891
800 meters 135.65±1.12 135.50±1.13 0.260 0.803
Standing long jump 2.83±0.04 2.84±0.03 -0.505 0.629
Solid ball 12.16±0.20 12.17±0.19 -0.110 0.915
Table 1 – four basic quality tests before the experiment
The students of experimental group and control group in the strength, speed, endurance
and the physical quality have no significant difference, in the 100 meters, standing long
jump and solid ball two project control group is still slightly better in the experimental
group. But in the 100 meters and 800 meters of the project, the control group, the relative
standard difference is slightly larger than the experimental group, and in the standing
long jump and solid ball project, controls the relative standard difference slightly light
rain in the experimental group.
4.2. Research results

In this experiment, because in the process of teaching, the experimental group and the
control group using respectively different training methods, the experimental group
with fixed line law practice, the control group using the traditional method of practice
and has a different effect of teaching. From table 2 show that after the experiment, the
two groups of students on the badminton ball technology learning interest has been
greatly improved. Students in the two groups of “no interest” choice is zero, indicating
RISTI, N.º E5, 07/2016 121

that badminton is well received by students welcome, also shows that whether it is
traditional gait practice or fixed line method step practice so that students will have
more love of badminton sport.
group Very interested General No interest

experience group 6 2 0
Percentage (%) 75% 25% 0%
control group 5 3 0
Percentage (%) 67.5% 32.5% 0%
Table 2 – Learning interest in badminton after the experiment
experience group control group

project P value
Before After Before After
100 meters 12.15 11.98 12.16 12.05 P<0.05
800 meters 135.65 128.52 135.50 130.21 P<0.05
Standing long jump 2.83 2.91 2.84 2.89 P<0.05
Solid ball 12.16 12.76 12.17 12.71 P<0.05
Table 3 – Students physical quality after the experiment
From table 3 it shows two groups after 24 hours of badminton class, on the level of
physical fitness have significantly improved, and in the experimental group and the
control group in the 100 meter run, 800 meter run, standing long jump, solid ball has
significant changes, the p value was less than 0.05. It is shown that there are significant
differences. And in these four basic qualities, the experimental group improved
significantly higher than the control group, the results show that the fixed line training
method has the actual effect, can be applied to the badminton teaching practice.
group Step technique Hitting technique

experience group 42.6 40.8
control group 38.5 36.9
Table 4 – Two groups technical level testing
As can be seen from table 4, the experimental group is not only the method of the
technical evaluation of the average score higher than the control group, and the
technical performance of the ball hit the average score is higher than the control group.
This shows that the fixed line method not only to make students in the process of the
122 RISTI, N.º E5, 07/2016

technical aspects of effective, but also has a great effect on the ball technology, but also
shows that the method of the direct impact of the technology. The results show that
the experiment is successful, has the actual effect and the utilization, can effectively
improve the student badminton ability level.
group Wins lost winning probability

experience group 6 2 75%
control group 2 6 25%
Table 5 – Results of teaching competition of the two groups of students
From table 5 can be seen: the students in the experimental group winning streak and
winning percentage significantly than that of the control group students is much higher,
which also fully illustrates the through 32 hours of study and practice, the technical level
of the students is improved, but the two groups of students to improve is differentiated,
scores of the experimental group was significantly better than the control group.
Analysis of the reasons, is due to the fixed line practice in teaching practice has played an
important role. Specific it, fixed line practice high density of the traditional badminton
practice methods, practice in the unit of time the number of relatively more, the load of
students is relatively large, but this can stimulate students interest in learning, the fixed
line law practice is more conducive to improve and enhance the quality of students’
action and rapid formation action to finalize the design.
5. Conclusions
Multimedia computer sports information processing system is more advanced multimedia
technology, which makes sports information processing system can effectively sports
information on a wide variety of comprehensive analysis and processing and will sound,
text and image data together, and stored in the database system of sports information
broadcast on the screen, or is the use of the network to transmit them. In addition,
the system also has interactive features. This feature is refers to the sports information
disseminators and recipients between real-time information exchange, and exchange
process is often people use system programming and dialogue, and through active
control system to carry out the work, in order to analyze, research has been accepted and
storage of sports information, make sure to really play the role of sports information.
Relative to the traditional badminton exercise, fixed line training method can enhance
students’ interest in badminton, the badminton, improve students’ heart and lung
function, enhance students speed endurance and indomitable fighting spirit. It can also
improve the students’ enthusiasm for the exercise of the badminton class. Fixed line
practice method on students’ Badminton Footwork technical ability raise has significant
effect, and can not only improve the students’ all aspects of physical ability and to
improve the quality of the badminton special also remarkable effect. The reasonable
control of the speed and direction of the students’ movement and the time between each
RISTI, N.º E5, 07/2016 123

group during the course of the fixed line method can bring different effects. The fixed
line training method can improve students’ initiative in Badminton Class and cultivate
students’ ability to innovate. The teaching of the fixed line training method has a good
effect on the students’ social adaptation ability, not only can cultivate students’ spirit of
solidarity and cooperation, but also can narrow the relationship between the students.
References
Crespo, P., Santos, V. (2015). Construction of Integrated Business Management Systems
for Micro and Small Enterprises. RISTI-Revista Ibérica de Sistemas e Tecnologias
Dai, W., Fan, L. (2012). Discussion about the Pros and Cons and Recommendations for
Multimedia Teaching in Local Vocational Schools. Physics Procedia, 33, 1144–1148.
Hu, M., Xu, S. (2012). Research of Multimedia Teaching on Principles of Management.
IERI Procedia, 2, 666–670.
Huang, Y., Backman, S. (2013). Experiencing student learning and tourism training in
a 3D virtual world: An exploratory study. Journal of Hospitality, Leisure, Sport &
Tourism Education, 13, 190–201.
Jian-hua, S., Hong, L. (2012). Explore the Effective Use of Multimedia Technology in
College Physics Teaching. Energy Procedia, 17, 1897–1900.
Kim, A., Davies, J. (2014). A teacher’s perspective on student centred learning: Towards
the development of best practice in an undergraduate tourism course. Journal of
Hospitality, Leisure, Sport & Tourism Education, 25, 6–14.
Krstev, C., Trtovac, A. (2014). Teaching Multimedia Documents to LIS Students. The
Journal of Academic Librarianship, 40, 152–162.
Sigala, M. (2012). Investigating the role and impact of geovisualisation and
geocollaborative portals on collaborative e-learning in tourism education. Journal
of Hospitality, Leisure, Sport & Tourism Education, 11, 50–66.
Zhang, C., Chen, X. (2012). Use of Multimedia in Gross Infective Pathogen Experimental
Teaching. Procedia Engineering, 37, 64–67.
124 RISTI, N.º E5, 07/2016

Study on the Cooperation of Hotels and Online

Self-service Travel Business based on Information
Technology
Xiaolong He, Xiaomei Zhang, Fangyuan Liu*
* 5499036@qq.com
College of Economics and Management, Northeast Agricultural University, Harbin 150030, China
Pages: 125–135
Abstract: With the rapid development of self-service travel and online travel
business, a new product called an “online self-service travel” business sprout, which
is an online platform providing the information of sightseeing, lodging, dining,
traffic, entertainment and shopping in tourism activities. Visitors could choose
the attributes and ways of travel matters according to personal preference, the
platform connects travel matters by itself, forms itinerary and shows related travel
information for visitors to accomplish self-service travel. This paper collected 191
hotels’ data from Harbin of Heilongjiang province, used Binary Logistic Regression
for studying the influence factors of cooperation of hotel and online self-service
travel business based on online tourism enterprise. At last, getting result through
SPSS, it shows that 12 influence factors are having positive effect such as tourism
enterprise reputation, technical guidance during cooperation, related policy
guidance and pricing method, 3 factors having no effect on the cooperation which
are hotel room quantity, tourism enterprise economic benefits and creativity.
Keywords: Online tourism enterprise, online self-service travel, hotel, cooperation
1. Introduction
Visitors are good at collecting kinds of travel information from online travel platform
for travelling in our era that tourism industry virtualization and network life get more
universal, especially young visitors like choosing self-service travel by using resources
and information from online travel platform (Zhenglu,2014; Delgado, 2015). Online
tourism enterprises and their online travel business came into being and could meet
parts of the demand of tourism market. Online travel business accompanied by the
online tourism enterprise, three stages has been underdone since the end of the 20th
century in China. The first stage from 1999 to 1999, it is the stages in germination
of the online travel business. Tourism enterprise obtained the market share through
the call centre or online booking in this period, and obtain the benefit, information
consultation and online booking is the main of online business (Siwei, 2014; Zhao,
2014; Lixin, 2015). The second phase from 2002 to 2007, is the evolution period of
the online travel business. E-commerce began to widely used during this period, to
combine the tourism boom and the electricity business boom, and play the role of the
RISTI, N.º E5, 07/2016 125

Study on the Cooperation of Hotels and Online Self-service Travel Business based on Information Technology
Internet automation, and then make the tourism enterprises to improve efficiency and
reduce cost, the kind of online business is increase with the need of market (Tao,2011;
James, 2009). The third stage from 2007 to now, is the mature period of the online
travel business. The application of smart phone as the start during this period, and
then mobile devices and mobile network obtained the widely used, and the APP mobile
application development is rapid, it is provided the new opportunities for tourism
enterprise, Tourism enterprises into the era of flowers contend gradually, to carved up
the tourism market demand using the multivariate model, and focus on one or more of
the online business (Yun, 2014; Wenbiao, 2014).
A small number of online travel companies such as Ctrip, to make a few main business
and stronger gradually on the basis of resources and experience advantages, and become
a comprehensive online business tourism enterprises; more online tourism enterprises
is focused on one or two, three types of business, such as the Donkey-mother travel
network focused on the custom services of tours and group tickets, and the Mafengwo
travel network focused on travel guides and user experience, and equivalent to the online
travel community.
At present, the online travel business exists in the Chinese market can satisfy the general
requirements of most visitors, but for many self-service visitors, they hope the tourism
market can appear an online travel business trip which the itinerary is choreography
according to the personal preference, and farewell the existing model which to find
all kinds of information of strange tourist destination, and design schedule through
repeated comparisons, This method saving part time on the one hand, and the content
be recommended with emphasis on the other hand.
2. The content of online self-service travel business

Because of situation that there isn’t an online business providing itinerary for visitors
according to their personal preference in tourism market, this paper put forward online
self-service travel business to fill the vacancies in the market. The online self-service travel
business is an online travel business platform for visitors based on one online tourism
enterprise including large information of travel scenic, hotel, traffic entertainment and
more, visitors according to personal preference choose the attributes and ways of travel
matters, the platform would connect travel matters by itself, form itinerary and show
related travel information for visitors to accomplish travel action, figure.1 shows the
business process.
The online self-service travel business is still in the prototype stage, at present, there
is no Chinese tourism enterprise to develop and launch on the market, this promote
more direction and space subdivision to study. As the hotel is one of the most important
partners for the business operation, and close cooperation with the relevant travel
business, hotel is regard as the main body of research; we can predict the future growth
space through the approval, acceptance and willingness of cooperation comes from the
hotel. What is noteworthy is that since the business practices is generated in the travel
business, unable to determine the cooperative profit distribution, the most concerned
issues of partner does not make the concrete discussion in this paper, it can regard the
average of profit distribution of the similar online travel business as a reference.
126 RISTI, N.º E5, 07/2016

Figure 1 – Business Process of Online Self-service Travel Business
3. System establishment of cooperation influence factors

Hotel having right to decide would suffer from a variety of factors while cooperation
between hotels and online tourism enterprises. This paper sorts out 4 classes and 17
factors of cooperation influence fitting city condition after finding out the present
situation of Harbin hotels and their cooperation with online tourism enterprises.
Therefore, the system showed in figure.2.
Specifically, 4 classes are aspect of hotel itself, aspect of online tourism enterprise, aspect of
the process of cooperation and other aspect. 17 factors are hotel type, hotel location, target
customer, room quantity, room occupancy rate, having self-marketing business and having
cooperation with tourism enterprise in aspect of hotel itself, online tourism enterprise scale,
economic benefits, reputation and creativity in aspect of online tourism enterprise, technical
guidance, cooperation method, pricing method and settlement method in aspect of the process
of cooperation, enterprise cooperation trend and related policy guidance in other aspect.
Along with the increasing influence of Internet and information technology to the
tourism industry, OTA (online travel service provider) and the majority of tourists
using the Internet spontaneously organized the “Lvyou group” are the traditional to
travel agency industry as representative of the travel service provider has had a huge
impact. In early twenty-first Century, with the popularity of the Internet, a new travel
provider OTA (online travel service providers) came into being. After nearly 10 years
of development, OTA in terms of revenue has been to catch up with the trend of the
traditional travel agency. According to the National Tourism Bureau statistics, in 2013
China’s travel agency industry annual revenue of 359 billion yuan, business growth rate
of 6.7%, but the domestic tourism and inbound reception declined. According to Ai
RISTI, N.º E5, 07/2016 127

Figure 2 – Cooperation Influence Factors System of Hotel with Online

Self-service Travel Business
Rui network statistics, in the same year, China’s online travel transactions amounted
to over yuan, an increase of 27.7%, among them, the online ticket market transactions
amounted to 131 billion.
The basis of traditional tourism intermediary service industry is the monopoly of
information, which is built on the basis of different cost of tourism intermediary service
industry and tourists to obtain information. Because tourism products involves many
aspects and has a remote and repeated use rate is low, and the characteristics, so the
traveler information collection cost is very high and is difficult to reuse, inhibit the normal
tourism initiative to collect information of power, and professional travel agency does not
exist this situation. With the rise of the Internet, travel product supplier for propaganda
purposes their information will be published on the Internet, users, especially after 70
after the birth of the habit to use network of ordinary tourists to collect information
related to the cost decline, and in recent years the rise of the UGC (user generated
content) and PGC (specialized in producing content) not only makes ordinary tourists to
128 RISTI, N.º E5, 07/2016

Figure 3 – The online self-help tourism become popular
collect relevant information costs decline further, but also improve the accuracy of the
information. With the rise of the Internet, the traditional intermediary service agencies -
travel agencies in the information on the dominant position of almost all loss.
With the development of the Internet, especially the rise of a number of tourism
professional websites, tourism product suppliers cannot through travel agencies directly
to the tourists before the product information, and its power will be sold to the lower
prices directly to the tourists, and to avoid the travel agency in the middle of the link.
Therefore, for ordinary tourists, not only the cost of information collection, transaction
costs also continue to decline, the price difference between the price of tourism products
with standardized. Even though the price of the quoted price is slightly lower, many
tourists may choose to travel freely because of the poor degree of freedom of the tourists
who buy the quoted price. From the current trend, the freedom to travel is becoming a
trend. Online booking hotel now is really convenient and simple, as shown in figure 4.
Figure 4 – Online booking hotel
RISTI, N.º E5, 07/2016 129

4.1. Source of data
A set of questionnaire conform to the content is designed according to the cooperation
influence factors system, the total question is 19, and the 17 influencing factors of
system are included. In order to ensure the accuracy of the results of the survey, the
questionnaire survey is a field survey, the specific respondents are deputy manager level
or above of large hotel and the hotel owner of small and medium-sized hotel, so ensure
that they can represent the policymakers of the hotel to participate in the cooperation.
Research out 210 questionnaires, recycling 209 questionnaire, valid questionnaires is
191, 90.95% of total questionnaire distributed.
4.2. Establish of model

Logistic regression model is nonlinear regression model of the probability, mainly used
in the regression analysis of qualitative variable; it is a multivariate analysis method to
study the relationship between the results of classification observations (Y) and influence
factors(x). The dependent variable can be more or binary classification. The research
content of this study is the hotel in Harbin to participate in online self-service travel
business based on online tourism enterprise or not, only “have” and “have not” as the
results of whether the hotel have willingness to cooperate, so binary Logistic regression
analysis model is the most appropriate.
Binary Logistic regression analysis model is:
 11

= (Yi 1=
pi p= X i ) F  α + ∑ β j X ij  (1)
 i 
In the above formula: “p” means a choice rate; “Xij” means factors; “F” means to submit
a standard Logistic distribution.The model is established according to the study:
(Yi 1=
p= Xi ) G =
(Z)
(
exp β 0 + ∑ i =1 β1 X i + µ
n
)

(
1 + exp β 0 + ∑ i =1 β1 X i + µ (2)
n
)
In the above formula, “p” means the cooperation probability of Harbin hotels choose to
participate in cooperation with online self-service travel business based on online tourism
enterprise; dependent variable “Y” means the willingness to cooperation of Harbin hotels
to participate in online self-service travel business based on online tourism enterprise,
“Yes” is expressed in “1”, “No” is expressed in “0”; “Xi” means all kinds of factors to
influence of Harbin hotels to participate in cooperation with online self-service travel
business based on online tourism enterprise, it is respectively for the hotel type(x1),
hotel location position(x2), hotel target customer(x3), hotel room quantity(x4), hotel
room occupancy rate(x5), hotel online self-marketing business(x6), hotel cooperation
with tourism enterprise(x7), tourism enterprise scale(x8), tourism enterprise economic
130 RISTI, N.º E5, 07/2016

benefits(x9), tourism enterprise reputation(x10), tourism enterprise creativity(x11),

technical guidance in the process of cooperation(x12), cooperation method(x13), pricing
method(x14), settlement method(x15), enterprise cooperation trend(x16), related policy
guidance(x17); “G” means cumulative distribution function of a standard Logistic
random variables; “u” means the random disturbance.
4.3. Result of data analysis

First, according to the above analysis, making dependent variable “Y” to variable
definitions, “Yes” is expressed in “1”, “No” is expressed in “0”.Second, according to the
content of every survey question option, making “Y” to variable definitions, the details
are showed in Table 1. At last, putting 17 explanatory variables into the model of Binary
Logistic Regression, this paper selected the statistical software is SPSS 19.0.
Variable Variable Range

x1(0 - five-star, four-star or luxury hotel, 1 - three-star or business hotel, 2 - 0~4
economic, express inn hotel, 3 - features inn, theme inn or youth hotel, 4 - other
accommodation hotel)
x2(0 - at or near tourist attractions, 1- at or near transportation hub, 2 - at or 0~5
near business center, 3 - at or near main urban roads, 4 - at or near school,
hospital, family area, 5 - other location)
x3(0 - mainly tourists, 1 - not mainly tourists) 0~1
x4(0 - less than 100, 1 - 100~400, 2 - more than 400) 0~2
x5(0 - less than 30%, 1 - 30%~50%, 2 - more than 50%) 0~2
x6(0 - have self-marketing business, 1 - have no self-marketing business) 0~1
x7(0 - have cooperation with tourism enterprise, 1 - have no cooperation with 0~1
tourism enterprise)
x8(0 - large scale, 1 - general scale, 2 - new year or small scale) 0~2
x9(0 - good economic benefits, 1 - general economic benefits, 2 - bad economic 0~2
benefits)
x10(0 - good reputation, 1 - general reputation, 2 - bad reputation) 0~2
x11(0 - have creativity, 1 - have no creativity) 0~1
x12(0 - have technical guidance, 1 - have no technical guidance) 0~1
x13(0 - book online and sale by itself 1 - book and sale on the Internet) 0~1
x14(0 - pricing by itself, 1 - pricing agreement with tourism enterprise) 0~1
x15(0 - single or one-day settlement, 1 - periodic settlement, 2 - irregular 0~2
settlement)
x16(0 - have influence, 1 - have no influence) 0~1
x17(0 - have influence, 1 - have no influence) 0~1
Table 1 – Variables definition of influence factors
RISTI, N.º E5, 07/2016 131

According to the established display form of Binary Logistic Regression result in SPSS
19.0, there are 5 tables show the result of data analysis. Table.2 shows on the “case”
description: 191 pieces of data involved in the calculation, there is no default.
Unweighted Casesa N Percent

Included in Analysis 191 100.0
Selected Cases Missing Cases 0 .0
Total 191 100.0
Unselected Cases 0 .0
Total 191 100.0
Table 2 – Case processing summary
Original Value Internal Value

Yes 0
No 1
Table 3 – Dependent variable encoding
Table.3 shows the result of encoding variable: the score of “1” represents “having willingness to
cooperate”, the score of “0” represents “having no willingness to cooperate”.
Predicted
Observed Agree to cooperate
Percentage Correct
No Yes
Agree to cooperate No 0 74 .0
Yes 0 117 100.0
Overall Percentage 61.3
Table 4 – Classification processing table

Table.4 shows that the correct rate of anticipating all the “case” having willingness to
cooperate without any arguments into the past, and the percentage correct:61.3 means
there are 61.3% of hotels involved in the investigation having willingness to cooperate
with online self-service travel business based on online tourism enterprise.
Predicted
Observed Agree to cooperate
Percentage Correct
No Yes
Agree to cooperate No 54 20 73.0
Yes 10 107 91.5
Overall Percentage 84.3
Table 5 – Classification processing table
132 RISTI, N.º E5, 07/2016

From Table 5, it shows that the accuracy in “case” classification by regression analysis,
generally prediction accuracy rate of 50% that is considered good. In this case the
percentage correct is 84.3 more than 50, that is to say classification in describing factors
are good. The table 6 shows the coefficient of variable of the output regression equation
and the test values for the coefficient, the Sig. less than 0.05 means the coefficient has
statistically significant, on the condition of this, B value is positive means the influence
of the independent variable on the dependent variable is positive, B value is negative
means the influence of the independent variable on the dependent variable is negative,
and the greater the numerical is, the greater influence. Table 6 is the final empirical
analysis result of the influence factors for Harbin hotels to participate in online self-
service travel business cooperation based on online tourism enterprise, it can be seen
from the table that there are 12 factors to influence the cooperation, and 3 factors has no
influence on the cooperation, the other 2 factors has no statistical significance. Summary
table could be seen in table 7. In the table, “+” represents the impact direction is positive,
and the more “+” are, the deeper influence is, “-” represents the impact direction is
negative, “——” stands for in this study the factor hasn’t get conclusion.
95% Confidence Interval for

factor B S.E. Wald df Sig. EXP(B) Exp(B)
Lower Bound Upper Bound
x1 .680 .311 4.774 1 .029 1.974 1.072 3.632
x2 .187 .206 .827 1 .045 1.206 .805 1.806
x3 .137 .969 .020 1 .888 1.146 .172 7.658
x4 -.768 .687 1.381 1 .040 .464 .129 1.671
x5 .763 .417 3.340 1 .032 2.144 .946 4.857
x6 .199 .654 1.005 1 .043 1.701 .817 4.323
x7 .909 .551 2.720 1 .035 2.483 .842 7.318
x8 .848 .629 1.606 1 .039 2.335 .629 8.672
x9 -.616 .580 1.130 1 .042 .540 .173 1.682
x10 4.063 .740 30.100 1 .000 58.123 13.616 248.115
x11 -1.648 .669 6.861 1 .017 .192 .056 .660
x12 2.029 .560 13.127 1 .000 7.604 2.538 22.785
x13 0.315 .610 .323 1 .047 1.370 .462 4.065
x14 1.598 .600 7.100 1 .008 4.944 1.526 16.021
x15 .241 .562 .184 1 .049 1.273 .423 3.831
x16 -.322 .555 .279 1 .571 .725 .219 2.394
x17 1.018 .614 2.747 1 .035 2.768 .830 9.224
Constant -12.534 3.021 17.216 1 .000 .000
Table 6 – Variables in the equation
RISTI, N.º E5, 07/2016 133

Factor Influence
Hotel type +
Hotel location +
Hotel target customer ——
Hotel room quantity -
Hotel room occupancy rate +
Hotel have self-marketing business +
Hotel have cooperation with tourism enterprise +
Tourism enterprise scale +
Tourism enterprise economic benefits -
Tourism enterprise reputation +++
Tourism enterprise creativity -
Technical guidance +++
Cooperation method +
Pricing method ++
Settlement method +
Enterprise cooperation trend ——
Related policy guidance ++
Table 7 – Empirical Analysis Result of cooperation influence factors
5. Conclusions
The results of survey data sorting shows that 191 Harbin hotels involved in the
investigation and there are 117 hotels has the cooperation intention, accounted
for 61.3%, it means that more than half of the hotels accept online self-help travel
business as a new tourism product, having the potential willingness and opportunities
to cooperation with it. Empirical analysis results show that hotel type, location, room
occupancy rate, having self-marketing business, having cooperation with tourism
enterprise, online tourism enterprise scale, reputation, technical guidance, cooperation
method, pricing method, settlement method and relevant policy guidance are the factors
to effect hotels to participate in the online self-service travel business based on online
tourism enterprise, and the influence of the most significant is online tourism enterprise
reputation, technical guidance and pricing method; but hotel room quantity, online
tourism enterprise economic benefits and creativity has no effect on the cooperation.
Based on the empirical analysis result, online tourism enterprise should establish a good
enterprise reputation at the same time of maintain their own economic development,
focus on the relevant policy guidance, pay more attention on the convenience of operation
on the design and promotion of online self-service travel business, give related technical
support in the process of enterprise cooperation, and give the hotel more initiative in the
process of pricing and settlement. At the same time, the hotel should be good at analyzing
the market situation, seize the opportunity, using the characteristics or advantages of its
134 RISTI, N.º E5, 07/2016

location and marketing channel to obtain more cooperation opportunities so as to stay

in business and expand the development. As the specific contents of the study online
self-service travel business is not existed in the tourism market, the range of the object of
study is limited. So the study results have certain representative but not comprehensive.
Future research could broaden cooperation fields of the content and expanding the
scope of the object choice.
References
Delgado, A. P., Velthuis, M. (2015). Proposal for a continuous improvement IT
governance framework at financial institutions. RISTI-Revista Ibérica de Sistemas
e Tecnologias de Informação, (15), 51–67.
James, H., Mark, W. (2009). Econometrics (the 2nd Edition), Shanghai: Gezhi Press,
294–295.
Lixin, Z. (2015). Research on development of economy hotels in Harbin. The Border
Economy And Culture, 4, 6–7.
Siwei, Z. Fang, F. (2014). Development of a model for a cluster-based virtual tourism
supply chain. Tourism Tribune, 29, 46–54.
Tao, C. (2011). Econometrics. Dalian: Dongbei University of Finance & Economics
Press, 211–216.
Wenbiao, L. (2014). Study on innovation online tourism enterprise business mode based
on supply collaboration. Logistics Technology, 33, 40–42.
Yun, L., Lingqiang, Z. (2014). The comparison research on business model innovation
paths of online travel companies. Science Technology and Industry, 2, 238–29.
Zhao, L., Sheng, H. (2014). Proficiency analysis of tourism enterprises’ network
marketing. Hebei Normal University of Science & Technology (Social Sciences),
13, 89–92.
Zhenglu, G. (2014). Analysis to the tourism demand of students and strategy to the
marking of tourism enterprises. Foreign Economic Relation & Trade, 11, 91–94.
RISTI, N.º E5, 07/2016 135

Prediction and Simulation on the Consumers

Purchase Intention of Rice based on System
Dynamics Modeling
Shengde Hu, Haoning Wu, Yuqiu Cai *
* 63cyq@163.com
Pages: 136–145
Abstract: The system dynamics can be applied to the medium and long term
economic analysis and forecast, as the problem of food security in China is a
complicated economic problem, so that SD method is an effective method to solve
this kind of complex problem. In this paper, we build the SD flow diagram of grain
supply and demand, and then research the consumers’ willingness on purchasing
rice based on binary logistic model. The result shows that consumer purchase
intention of rice is influenced by gender, place, market concern degree, quality
and safety, price range and safety knowledge. Rice as a consumer’s necessities, is
different from other agricultural products, consumers have a higher level of quality
and safety requirements on the rice consumption.
Keywords: System dynamics, causality diagram, purchase intention, rice products
1. Introduction
China’s government for years to the formulation and implementation of various
policies to protect and promote the stable development of grain production, but in the
food policy also had some mistakes, such as policy targeted poor, failing to anticipate
changes in supply and demand, the indirect effect of the policy and the long term effect
of estimation problems and so on (Joshua, 2015; Delgado, 2015; Milad, 2015). Past of
traditional experience judgment and qualitative analysis of agricultural policy analysis,
has been far cannot meet the new situation policy and economic development demands
more scientific, systematic and quantitative policy analysis tool, study of agricultural
policy in China has gradually from qualitative analysis to quantitative analysis. At the
same time, in recent years, the frequent occurrence of food quality and safety problems
caused by the strong reaction of the consumer group, which has a lot of scholars on
the issue of food safety and consumption (Ali, 2014; Gopal, 2014;Natascha, 2015). For
consumers of stratified sampling survey, through descriptive statistics and cross project
analysis method to measure consumer purchase intention of on food safety based, verify
the food safety information can be fortified to increase consumer’s willingness to buy.
High quality rice industry under the joint efforts of relevant government departments
and rice enterprises, there has been a considerable development, Harbin Wuchang,
136 RISTI, N.º E5, 07/2016

Mudanjiang and other well-known high-quality rice producing areas of rice products on
the market price has been much higher than the average market price of the common
rice, but most of the country’s rice production has not yet been completed by ordinary
rice to the transformation of high quality rice, rice products on the market price is not
high, even in accordance with the standard production of high quality rice products, it is
difficult to obtain the consumer recognition, rice products of good quality and low price
to achieve a long-term and arduous task(Ilyoo, 2013; Aurelio, 2013). If it is difficult to
achieve higher prices for better quality, will inevitably affect the rice farmers planting of
high quality rice enthusiasm, make the quality and safety of products of rice is not stable
supply, influence of our country food security. The realization of high quality and low
price of rice products is not only a process of changing from “lemon market” to “high
quality market”, but also an important means to improve the income level of farmers and
processing enterprises. Therefore, it is necessary to the realization path of rice industry
high quality construction, realize the production, processing and marketing links closely
connected, of industry in the aspect of the subject of interest coordination and stable
and good quality rice production, meet the needs of consumers for the purpose of the
demand of high quality rice.
2. System dynamics modeling

System dynamics to build in control theory, system theory and information theory based
on the study of the feedback system structure, function and dynamic behavior of a kind
of method for long-term social economic analysis and prediction, known as the “strategy
and decision Lab. Professor Wang Qi fan in the early 1980s introduced the system
dynamics method into China, has triggered a hot SD, in recent years, system dynamics
application of growth trend once again has become one of the hot topics in the study of
complex systems. Change law of China’s food security is a complex economic problems,
many of the details we did not fully grasp the rules and inherent relation, for this kind
of complicated problem of SD method is an effective method to solve the problem. In
this paper, the system dynamics method is used to describe the relationship among
the variables, and simulate the change law of the system, which is used to analyze the
problem of grain, to realize the medium and long term forecast and policy simulation.
2.1. System factor analysis

The basic balance of grain supply and demand means that there is enough grain supply
to meet the demand of grain. The grain supply capacity of a country or region depends
on the amount of grain production, the amount of imports and the beginning of the
period, the demand of grain depends on the amount of consumption, the export volume
and the final reserve quantity. Total food production is jointly decided by the per unit
area yield of grain and grain sown area, respectively, on Grain Yield and grain sown
area were analyzed: there are many factors affecting grain yield per unit area, generally
speaking, are mainly six aspects: labor input, agricultural production inputs, the
promotion of agricultural science and technology progress and technology, National
Water Conservancy Agricultural Public Investment in infrastructure, related system and
policy and natural disasters influence; grain sown area is mainly composed of arable land
RISTI, N.º E5, 07/2016 137

Prediction and Simulation on the Consumers Purchase Intention of Rice based on System Dynamics Modeling
resources, multiple cropping index and grain planting area accounts for crop acreage
proportion of three aspects.
Based on domestic resources to achieve self-sufficiency in grain, is the basic policy to
solve the problem of grain supply and demand in china. The State Council issued the
“China’s grain problem” white paper pointed out that under normal circumstances
China’s grain self-sufficiency rate of not less than 95%, net imports not more than 5%
of domestic consumption. Previous experts have suggested that the self-sufficiency
rate is about 90%, taking into account the 10% of China’s imports of food will lead to
fluctuations in the international market food prices, and thus affect the third world food
supply. In the medium and long term prediction, the aggregate supply and demand for
food is the focus of our attention, and food supply and demand situation affects the food
price changes, price changes and further affect the next period of grain production.
2.2. Causality diagram (CLD)

Considering the purpose of modeling, analysis of the information collected from various
elements of abstract description of system overview and representative variables,
establish the system variables: the total grain yield (Total Production), yield (Production
per Unit), labor input (Labor), the total power of agricultural machinery (unit, Machine)
per mu per mu the amount of chemical fertilizer (unit Fer), the effective irrigated area
(Irri Area), water conservancy investment accounted for the national infrastructure
investment proportion (Irri Rate), agricultural science and technology three accounted
for the proportion of fiscal expenditure (AgriTech/GovExp), the disaster area ratio
(DisRate), grain sown area (Planting, Area) (Plant Area), cultivated land area multiple
cropping index (MI), the sown area of grain sown area of crops accounted for the
proportion of (Grain/Crop Area), the total food consumption (Consumption), total
population (total population), per capita food consumption (consumption per person),
inventory (stock), inventory coefficient (Stock Co), expectations (expected stock
inventory, expect (expected price) prices, grain price change rate of food price change,
prices and grain price), net import amount (net import).
Figure 1 – Causality diagram of the grain supply and demand
138 RISTI, N.º E5, 07/2016

According to the analysis of the main influence factors and determine the set of variables,
which can be used to build the system the main feedback loop: total production
arrows↑⇒Stock arrows↑⇒ StockCo arrows↑⇒ Expected price↓⇒Food price
change↓⇒Grain/CorpArea↓⇒Planting area↓⇒Total production decreases. This
circuit said food production caused by inventory increase, which leads to the reduction of
the expected prices and grain sown area reduced, leading to a period of food production.
Based on the above, the causality diagram is drawn, as shown in Figure 1. The variables
that affect grain production can be quantified by the variable of black body marker.
2.3. SD flow diagram

Causality diagram can only be qualitatively a rough description of system feedback
mechanism, if to quantitatively describe and shall be Figure 1 shows the causality
diagram change plotted SD flow chart, as shown in Figure 2, and on this basis, the use of
SD special dynamo language writing equations.
Figure 2 – SD flow diagram of grain supply and demand situation
3. System simulation and analysis

3.1. Simulation results
Application of system dynamics modeling software Vensim PLE model of food supply
and demand situation in China from 1996 to 2030 the time span of 35 years of simulation,
the main output results as shown in Figure 3, figure 4 and figure 5 shows. As can be seen
from the simulation results, the increase of grain yield in China depends on the increase
of yield, which is mainly caused by the decreasing of cultivated land in China. The next
two years, thirty years, China’s grain production and consumption tends to be stable,
food self-sufficiency rate of 93% to 107% fluctuations. From a long-term point of view,
in food production to keep steady growth, China’s food price volatility is very small,
compared with the prices of other consumer goods, prices steadily trend.
RISTI, N.º E5, 07/2016 139

Figure 3 –Food prices situation simulation curve
Figure 4 – Food production situation simulation curve

System dynamics method is mainly through the analysis of the structure of the social
economic system, with the help of computer simulation technology research, known as
the actual system of the laboratory. In the model, the policy parameters take different
values, which can be considered as different simulation schemes. The parameters in
this paper are based on the historical development of the system and the experience of
experts. In fact, these parameters can be adjusted with the real changes of the national
policy, which makes the model more realistic and effective. The simulation model itself
is a kind of abstraction and simplification of the realistic problems, due to a lot of data
that are difficult to quantify, coupled with the difficulties of data collection, the SD model
is constructed in this paper without considering agricultural water shortage, for farming
land quality, international prices of the impact and our government implementation
140 RISTI, N.º E5, 07/2016

Figure 5 – Total output of grain situation simulation curve
grain food purchase price policy and other factors. In addition, system dynamics model
is mainly used for quantitative analysis of long-term, dynamic and strategic. It focuses
to study the dynamic behavior of a system that is more concerned about the trend
prediction rather than the accuracy of the simulation results.
4.1. Econometric analysis model
This study to consumers to buy quality and safety of rice (the quality and safety of
rice is to reflect the quality and safety for the purpose, according to the organic food,
green food and pollution-free food standard production of organic rice, green rice, no
environmental pollution of rice in general, also known as for rice, the main products of
known for the quality and safety of rice or rice safe) will as explanatory variables, and
will be interpreted variables were transformed into binary variables (i.e. rice consumers
positive with “1”, the other with “0”) is a 0/ 1 binary variables. In the research of social
science, the general multiple linear regression models cannot be established directly
when the explanatory variables are 0/1 two variables, and most of the two Logistic
models are used in the study. Therefore, this study uses two yuan Logistic model,
which is interpreted as a variable range of 0-1. When the explanatory variable is the
two categories, the explanatory variable y is subject to two distributions,The overall
probability of setting the y=1 is P (=1 y),The probability distribution function of Y is:
(1 − y )
( y ) P Y (1 − p )
f= (1)
Under such a definition, the parameters of the model can be estimated by the maximum
likelihood estimation method. Log likelihood function:
n
( β ) log l=
l= (β ) ∑{y log 1 − F ( x ′β ) + (1 − y ) log F ( − x ′β )} (2)
i =0
i i
RISTI, N.º E5, 07/2016 141

The Logit model obeys the logistic distribution:
e xu β
( yi 1=
p= xi ) (3)
1 + e xu β
General form of the Logit model is as follows:
 m  1
G  α + ∑ β j X ij  =
pi =
 1 + exp −  α + β X  
 m

∑
j =1
  j ij  
  i =1 
4.2. Sample description statistics

Data from large supermarkets in Heilongjiang, residents of the area and part of the
province of rice consumption more area investigation and. Details of the survey involved
in the investigation of the family of basic information, family income, the purchase of
safety awareness of rice and the safety of rice consumption behavior and other aspects
of the detailed information. After several revisions, the author issued a total of 450
questionnaires, 421 questionnaires were recovered, the examination does not conform
to the logic of the sample 53, a total of 368 valid questionnaires, the effective rate of
the questionnaire reached 81.8%. The statistics of the basic survey sample are shown
in Table 1.
Item Max Min Average Standard Median

deviation
Quality safety 1 0 0.64 0.48 1
Gender 1 0 0.51 0.50 1
Age 4 1 2.63 0.83 3
Educational level 3 1 2.29 0.80 3
Price 5 1 2.95 1.16 3
Quantity 4 1 2.53 0.76 2
Place 5 1 2.67 1.11 3
Package 3 1 2.24 1.16 2
Concern 6 1 3.38 1.46 4
Price assessment 1 0 0.55 0.50 1
Price increase reception 3 1 1.76 0.62 2
Security rice 4 1 2.48 1.04 3
Consumption influencing factor 6 1 2.27 1.01 2
Table 1 – Statistical value of rice consumers′investigation characterization

At this stage of the rice market in the face of frequent emergence of rice quality and
safety problems, this study designs the rice consumers of rice and the suffering of the
market survey, in the “buy rice, worry about whether rice quality and safety,” the survey,
44.6% of consumers said worry, 55.4% of consumers said not very worried, that most
142 RISTI, N.º E5, 07/2016

consumers still has not formed the rice quality and safety consciousness of suffering.
At present, the price of rice in the rice market is generally low, and consumers in the
survey also think so. In the current price of rice on Evaluation of the survey, 34.2% of
consumers believe that the market the price of rice is low, 56.0% of consumers believe
that rice prices are moderate, only 9.8% of consumers rice prices are too high, is outside
the acceptable range. For at this stage, the vast majority of consumers think that the
price of rice bear in the range, consumer survey for the quality and safety of rice paid
to afford the highest prices, the design problem of a price increase to receive the degree
of. The results showed that the quality of the safety of rice was higher than that of the
ordinary rice price range <50% consumers accounted for 80.2% of the total sample.
Consumers have an important role in the safety of rice consumers to choose the quality
and safety of rice consumers. Therefore this study of consumers on rice understanding
cognition degree and the quality and safety of rice cross analysis, the results of pollution-
free, green, organic rice, the more understanding of consumers that is in line with the
“Sanpin” standard is the quality and safety of rice. In the “buy rice, how to identify the
quality and safety of rice, 32.6% of consumers think” brand rice is the rice quality and
safety, brand effect to consumers with a suggestive role, 26.1% of consumers use look
and comparison of “goods than traditional way” to choose, also labels indicate has a
more important effect on consumers.
4.3. Estimation results

In this study, the logistic regression analysis was conducted to investigate the data of 368 rice
consumers by using SPSS16.0 statistical software. The results of the model show that the log
likelihood function of -2 times is 186.423, and the numerical value is smaller. At the same
time, the NagelkerkeR2 is close to 1, so the model fitting degree is higher. In data processing,
all of the explanatory variables are adopted in this study. After repeated adjustment and
selection, the model is analyzed and the following results are obtained (Table 2).
Standard Wald
Item Coefficient Significance Occurrence
error outcome
Ⅰ Gender 4.788*** 1.306 13.873 0.000 1.297
Place of
-1.926*** 0.415 21.547 0.000 0.146
purchase
Ⅱ
Purchase
0.639*** 1.405 20.711 0.000 0.002
concerns
Market Worry
0.089* 0.169 3.765 0.060 1.093
Degree
Ⅲ Increase
acceptance -0.825*** 1.178 49.110 0.000 3.838
range
What is the
Ⅳ quality and 0.614*** 1.485 17.103 0.000 0.002
safety of rice
K constant 29.074*** 5.741 18.803 0.000 2.584
Table 2 – Result of quality rice consumption willingness model
RISTI, N.º E5, 07/2016 143

The influential factors for the individual characteristics of consumers in the consumer
to consumers to buy rice quality and safety has the positive effect of the 1% significance
level, that female consumers to buy rice pay more attention to the quality and safety of
rice. Female consumers because of the unique serious, meticulous inner characteristics,
items on the requirements are relatively high; at the same time, the majority of female
consumers as the main person in charge of the family diet, on the health of family
members has a more stringent standards and requirements, so buy rice meters will
pay more attention to the quality and safety of rice, which is in line with consumer
behavior rationality. In the influencing factors of the consumption characteristics, the
attention points of rice buy place and consumers buy rice have a significant impact.
Rice consumers purchase place on consumer purchase behavior of quality and safety
of rice had a significant negative to influence by the 1% significance level, indicating
that consumers consider the brand of rice shops selling rice quality and safety level is
higher, were more likely to buy quality and safety of rice. When consumers buy Rice’s
concerns are significant at the 5% level, indicating that the purchase of rice on quality
and safety information to consumers more willing to purchase rice quality and safety.
In the market environment factors, the sense of the market through the significant
level of 10%, the quality and safety of rice price increases by 1% of the significant level.
Consumers are aware of the quality and safety of rice in the existing market, but also
hope to find the quality and safety of rice from the level and type of hybrid market.
Quality and safety of rice to accept the magnitude of the increase in consumer purchasing
quality and safety of rice will decline in the behavior of rice. From the result of the model
can be seen, price increases an amplitude variable coefficient is negative, that with
the quality and safety of rice prices relative to the common rice rise amplitude bigger
inhibition of consumer purchase behavior of quality and safety of rice, which is in line
with the economics principle of elasticity of demand. Quality and safety of rice is not
only to maintain the effectiveness of the life, but also to enable consumers to have a more
healthy body, more energy to engage in social activities, has a strong external effect.
Therefore, consumers on the quality and safety of rice to understand the more detailed,
the more they want to buy quality and safety of rice. At the same time, the accuracy of the
understanding of quality and safety of rice, will better guide consumers to buy quality
and safety of rice.
5. Conclusions
This study based on consumers’ purchase willingness of rice quality and safety and its
influencing factors empirical analysis, the conclusions are as follows: rice consumers to
buy rice quality and safety will affected by gender, rice place to buy, buy rice attention,
degree of the suffering of the market, quality and safety full price for rice to accept
the influence of amplitude and the quality and safety of rice degree of understanding
factors. Different factors influence and affect the direction of each in a different way.
Specifically, women often in franchise stores to buy rice, buy rice attention to safety
and quality, with strong rice market consciousness, can bear larger quality and
safety of rice price accepted range, on the quality and safety of the rice to understand the
profound and clear understanding, has the characteristics of consumers to buy quality
and safety of rice a greater willingness. Rice as a consumer’s necessities, and other
144 RISTI, N.º E5, 07/2016

agricultural products, consumers have a higher level of quality and safety, more stringent
requirements. According to the above analysis, the author puts forward the following
suggestions: first, according to the consumer characteristics of individual characteristics
and location, to buy rice points of concern to the rice market segmentation, develop
differentiated marketing strategy of the quality and safety of rice; second, starting from
the consumer awareness and the quality of the price point of view, should strengthen the
supervision of rice market, beware of fake the security market is flooded with fake rice,
the implementation of high quality strategy, in order to foster harmonious rice market;
third, through various publicity channels, shaping the quality and safety of rice quality
image, so that consumers are fully aware of the ecological value of the quality and safety
of rice, in order to improve the quality and safety of consumers to buy rice willing.
References
Ali, I., Fauziah, S. (2014). Intercultural Communication Competence as a Key Activator
of Purchase Intention. Procedia - Social and Behavioral Sciences, 15, 590–599.
Aurelio, G., Mauri, R. (2013). Web reviews influence on expectations and purchasing
intentions of hotel potential customers. International Journal of Hospitality
Delgado, A. P., Velthuis, M. (2015). Proposal for a continuous improvement IT
governance framework at financial institutions. RISTI-Revista Ibérica de Sistemas
e Tecnologias de Informação, (15), 51–67.
Gopal, D. (2014). Factors affecting Indian shoppers’ attitude and purchase intention: An
empirical check. Journal of Retailing and Consumer Services, 21, 561–569.
Ilyoo, B., Hoon, S. (2013). The mediating role of consumer trust in an online merchant in
predicting purchase intention. International Journal of Information Management,
33, 927–939.
Joshua, D. Newton, Y. (2015). Environmental concern and environmental purchase
intentions: The mediating role of learning strategy. Journal of Business Research,
68, 1974–1981.
Milad, D., Mustafa, T. (2015). A research on effectiveness of Facebook advertising on
enhancing purchase intention of consumers. Computers in Human Behavior, 49,
597–600.
Natascha, L. (2015). The effect of food shape abnormality on purchase intentions in
China. Food Quality and Preference, 40, 24–30.
RISTI, N.º E5, 07/2016 145

Research on Evaluation Index System of Tourism

Competitiveness based on Tourism E-business
Platform
Wu Ling, Li Long
wuling2021@163.com

Pages: 146–157
Abstract: With the development of the tourism network platform, tourism

construction is a long-term and systematic work, the construction of evaluation
index system of tourism is very important. In this paper, the author made some
summary on tourism strong province status, and then construct evaluation index
of tourism competitiveness by using literature information method. On the basis
of analyzing the factors influencing the tourism construction, and followed the
principle of objectivity, systematic, measurability and comparability, the author
construct tourism evaluation index, including five primary indices as tourism
market needs capacity, tourism supply capacity, tourism support force, tourism
guarantees force and tourism pull effect.
Keywords: Tourism network platform, tourism competitiveness, evaluation
index system, industry contribution
1. Introduction
The national tourism administration put forward to build a World Tourism Power,
the first in the 2000 fifteen national tourism development plan, the next 20 years, the
goal of building a world tourism power in China will be able to achieve in this historical
period. In the national tourism work conference, first put forward the struggle goal
that we should transferred China from the world tourism power to become the world’s
tourism strong country (Hongjian,2009; Claudia, 2013). In 2005, the National Tourism
Administration has set up a special research topic, which is the first time to systematical
study the connotation and index system of China’s construction of the world tourism
power. In 2009, the State Council document No. forty-one of “the opinions of the State
Council on accelerating the development of the tourism industry” further clarified the
three aspects of this goal is the industrial scale, quality and efficiency. Efforts to reach the
time node is 2020, Strive to achieve the basic level of the world tourism power of China’s
tourism industry scale, quality and efficiency in 2020. In 2015 the national tourism work
conference, Li Jintao representative of the National Tourism Administration: the next
35 years, China’s tourism industry will be large to strong, from fast to good, to achieve
from the “preliminary well-off country” to “comprehensive well-off tourism country”.
146 RISTI, N.º E5, 07/2016

Then to the “preliminary rich tourism power” of the new leap, the next 35 years to build
China into world tourism power timetable and roadmap (Yongjun, 2008; Lijing, 2010).
Since the National Tourism Administration proposed the development goal of building
a tourism powerful country, many provinces put forward building the tourism
strong province, especially the abundant tourism resources provinces, which actively
explore effective policies and development strategies of the tourism strong province
construction(Zhang, 2014). In early 2015, the Implementation opinions of the Anhui
Province People’s Government on promoting the reform and development of tourism was
promulgated, which put forward that by 2020, the province’s total tourism consumption
amounted to 8000 billion Yuan, added value of tourism accounted for the proportion
of GDP of the province and strive to reach 7%, basically completed the strong province
of the travel (Zhifa, 2011; Xiangyang, 2009). On 7 August 2015, Anhui Provincial
Tourism Bureau in the document on the construction of strong tourism province to
build a comprehensive views of happiness, beautiful, wisdom and credit tourism,
which mentioned the four aspects of: happy tourism, infrastructure, beautiful tourism,
boutique, wisdom travels, Internet and tourism, tourism credit, credit information
publicity (Song,2005; Zhang, 2007). The construction of tourism strong province is a
major strategic decision made by the Anhui Provincial Committee and the provincial
government, in order to adapt to the new normal and promote the new development,
which is the stage goal of tourism reform and development (Qiuyang, 2009; Li, 2014;
Freixo, 2014). To create a comprehensive four tourism is key tasks and basic path for
the construction of the strong province of the travel. Building the strong province of
the travel, they pointed out the objectives and tasks of reform and development of the
tourism industry in the province; draw the outline of the overall framework, given the
direction and power of our province(Sá & Rocha, 2012; Abraham, 2000).
However, an important foundation for the construction of tourism strong province,
tourism industry and tourism product development still exist some problems, especially
tourism industry development is not balanced, tourism products obsolete has become an
important bottleneck of Anhui tourism strong province construction and the realization
of strategic targets. Therefore, in the basics of analyzing factors that influent the
construction of tourism strong province, how to construct the evaluation index system
of tourism province, to conduct a comprehensive assessment of the tourism strong
province, becomes the focus of our study.
2. Evaluation of online travel platform competitiveness

2.1. Tourism E-business Platform
From the development trend of the global tourism market, online travel industry has
gradually become a new industry. Online travel products derived from the ticket, hotel,
travel and other forms of tourism products. At present, the online travel business model
showing a diversified, differentiated momentum, has formed a complete upstream and
downstream industry chain, the enterprises in the competition continues to change, the
industry will present the pattern of competing patterns. Online travel platform model as
shown in figure 1.
RISTI, N.º E5, 07/2016 147

Research on Evaluation Index System of Tourism Competitiveness based on Tourism E-business Platform
Figure 1 – Online travel platform model

Tools for exploring deep social relationships will no doubt provide value for users and
businesses using Facebook. In the field of tourism, the user can easily find the next
journey companion and travel advice, had recently travel friends, as show in figure 2,
in the tourism industry of friends and friends of friends are those people are long-term
assignments in local and can provide travel advice. The long list of information provided
by Facebook will allow travelers to more fully understand the relationship between
them and the global social network. Users can also search for other points of interest,
including movies, books and music, they can search and use the Internet users to their
friends all the interest in public information. At the same time, mobile phone software
of online travel is also a new trend.
Figure 2 – Facebook search for social travel companion
148 RISTI, N.º E5, 07/2016

Figure 3 – Mobile phone online travel platform
2.2. The evaluation of the competitiveness of tourism

Many scholars believe that the concept of tourism strong province should be to define
from the competitive angle; the evaluation of tourism strong province also should follow
the ideas of competitiveness evaluation to construct. Therefore, the research on the
competitiveness evaluation method is the premise of constructing the evaluation index
system of tourism province. At present, the international recognition of the competitiveness
of the evaluation method is the world economic forum announced the travel and tourism
competitiveness index (TTCI) system and WTTC tourism competitiveness indicator.
Gooroochurn and Sugiyarto according to the method, use the WTTC, the World Bank,
and the United Nations figures released in 2002. With the help of AMOS4.0 software, the
tourism competitiveness of 93 countries and regions in the world was measured, found
that the United States, Sweden, Norway, Finland and Australia’s strongest competitive
destination, these countries generally have higher GDP per capita, value-added service
and trade rate. Tim Padmore and Hervey Gibson improve the diamond model of Professor
Potter; he put forward a kind of 3 factors model of industrial cluster competitiveness
as Groundings - Enterprise - market model (GEM). GEM model is the world of cluster
competition standard to 6 factors, evaluation of the competitiveness of industrial clusters.
The top brands of China’s online travel platform as shown in figure 4.
Geoffrey think that the main influence factors of tourism destination competitiveness
is a supporting resources and facilities, core resources and attractions, destination
management, destination planning and development policy, restriction and
amplification factors and so on, and based on porter’s diamond model theory, put
forward the “conceptual model of tourism destination competitiveness”, namely Crouch
& Ritchie model. Metin and Rimmington put forward a set of evaluation system of
Tourism Destination Competitiveness, based on the combination of qualitative and
quantitative indexes. They think the qualitative indicators including tourism like most
or least favorite tourism destination attributes and activities project and so on, it called
“soft” date. Dwyer & Kim model think there are a lot of destination competitiveness
indicators, both subjective attribute, also includes the objective attributes, destination
RISTI, N.º E5, 07/2016 149

Figure 4 – China’s online travel brand platform
market segmentation and tourism foreign exchange income. National or regional social
economy attribute index related to the main macroeconomic production, including
economic productivity levels, the overall level of employment, income of per person,
economic growth and so on.
3.1. Tourism market demand ability index
Based on the analysis of domestic and foreign literature, combined with the development
situation of tourism industry in Anhui Province, by brainstorming, the preliminary
conclusion is that the impact of tourism in Anhui Province construction factors, including
internal factors and external factors. Among them, the external influence factors are
mainly environmental factors, including social and economic factors, policy and legal
factors, scientific and technological progress factors, and so on. Internal impact factors
are various, mainly including supply and demand factors, specific tourism resources,
tourism location factors, human resources, capital supply factors, tourism market
factors, consumer demand factors and other factors, as shown in Table 1.
Environmental factor Social and economic factors, policies and legal

factors, scientific and technological progress
factors
Tourism Province
Supply factors Tourism resources, tourism location factors,
influential factors
human resources, capital supply factors, etc.
Demand factor Tourism market factors, consumer demand
factors
Table 1 – The influence factors of tourism strong province
150 RISTI, N.º E5, 07/2016

After several rounds of brainstorming, combined with the use of nominal group and
Delphi method, research group ultimately determines the tourism strong province
construction evaluation index, including five layer system elements: tourism market
demand capacity, tourism supply capacity, tourism support force, tourism security and
tourism pulling effect, which is the first class of tourism province evaluation indexes.
In order to establish the index system of evaluation index and the index factor, the task
group has carried out the further discussion to draws the following conclusions.
Tourism market demand ability is the most direct system factors that affect the
development of regional tourism industry, mainly includes the number of tourist
consumption, tourism consumption level, tourism consumption behavior, tourism
consumption potential and other factors (as shown in table 2). Tourism consumption
behavior related to the structure of tourism consumption. Among them, day tourists
per capita consumption level and the average length of tourists stay directly affects the
income of the local tourism, and the revisit rate reflects the source countries or regions
people travel frequency of tourism demand scale and ability.
System layer A System layer B System layer C

Number of domestic tourists
Number of tourist consumption
Number of foreign tourists
Domestic tourism income
Tourism consumption level
Foreign tourism income
Per capita consumption of tourists
Tourism market
demand Tourism consumption behavior Average length of stay
Revisit rate
Development potential of domestic tourism
market
Tourism consumption potential
Development potential of foreign tourism
market
Table 2 – Tourism market demand ability index
3.2. Tourism supply capability index

Tourism supply ability mainly includes the tourism resources endowment, tourism
product development, tourism enterprise development potential, tourism professional
talent supply potential and other factors (as shown in Table 3). Tourism resources
endowment is the basis of regional industrial development, directly determine the
tourism supply capacity. The ability of tourism product development determines
the development scale of regional tourism industry. Tourism enterprises are the carrier
and driving force of regional tourism industry development; tourism professionals are
the blood of regional tourism industry development and the supply potential of tourism
professionals play a very important role in the innovation and development of regional
tourism industry.
RISTI, N.º E5, 07/2016 151

Quantity and characteristics of tourism resources
Popularity
Tourism resource Rare degree
endowment
Grade degree
Degree of development
Product features
Product brand
Tourism product
Tourism development Product attraction
supply
capacity Additional product
Total tourism income
Tourism enterprise tax

The development of tourism Total labor productivity of tourism enterprises
enterprises potential
Number of travel agencies
Star Hotel quantity
Talent supply potential of The quantity and quality of tourism colleges

tourism professionals The number of personnel engaged in tourism
Table 3 – The index of tourism supply capacity
3.3. Support force index of tourism strong province

The strong province of the travel support force mainly includes macro-economic
development level, the attitude of the government and management ability, cultural
development and scientific and technological innovation elements (as shown in Table
4). The macro economic development level is a measuring instrument that reflects the
development of the regional economy; it is also an internal support for the development
of regional tourism industry. Government attitude and management ability is also an
important factor affecting the development of regional tourism industry, mainly related
to the government’s credit information publicity, financial revenue and expenditure
status, the degree of legal policy, the tourism industry’s capital investment, the level of
industry management, publicity and other aspects of the factors. Cultural development
and scientific and technological innovation can not only enhance the taste and cultural
connotation of regional tourism industry, but also improve the scientific and technological
content and competitiveness of regional tourism industry.
152 RISTI, N.º E5, 07/2016


GDP Per capita
Tourism added value accounted for GDP
Macro-economic
Tourism Investment
development level
Total consumption of social retail goods
Engel coefficient of urban residents
Credit information publicity
Tourism
industry Government attitude Balance of financial revenue and expenditure
support and management Degree of perfection of legal policy
force ability
The level of capital investment in tourism industry
Publicity and promotion efforts
The number of students in all kinds of schools
Cultural development The per capita funding for science and education
and innovation of The number of scientific and technological innovation
science and technology of tourism every years
The construction level of tourism information system
Table 4 – Support force index of tourism strong province
3.4. Security index and pulling effect index

The tourism security force mainly includes infrastructure construction, location,
environmental carrying capacity and other factors (such as shown in Table 5). Tourism
environmental carrying capacity is an important indicator to determine the development
potential of regional tourism industry, the tourism environmental carrying capacity is
analyzed and studied, which can guide people to reasonable management of tourism,
control the flow of tourists, prolong the life cycle of tourism products, and realize the
sustainable development of tourism industry. Tourism pulling effect is a measure of
the regional tourism development on the contribution of the regional economy and
the development of the multiplier effect of the tourism industry. The design of tourism
industry contribution factor specific targets is mainly from a consideration of the six
elements of tourism, the tourism food, housing, travel, shopping and entertainment
six elements of regional economic contribution degree basically reflect the regional
tourism industry to regional economic contribution degree (as listed in Table 6).
Traffic line density
Toilet density in tourist area
construction of infrastructural
facilities Number of tourist facilities owned by
thousands of people
Tourism facilities density
The tourism security Location advantage
Location condition
force Transportation condition
Ecological environment carrying capacity
Tourism resources carrying capacity
Environmental carrying capacity Carrying capacity of community residents’
economic life
Psychological carrying capacity of community
residents
Table 5 – Security index of tourism strong province
RISTI, N.º E5, 07/2016 153


Tourism revenue accounted for GDP
Tourism food and beverage contribution
Industry contribution Tourism accommodation contribution

Contribution of tourism and recreation
Tourism pulling effect
Tourism shopping contribution
Contribution of tourism traffic
Per million people engaged in the number of
Employment multiplier tourism related industries
The number of jobs created by tourism
Table 6 – Tourism pulling effect index
4. Evaluation index system construction

After defining the factors that influence the construction of tourist province, we need
to establish a complete set of evaluation index system, carries on the analysis and
evaluation of the construction degree of tourism strong province. Because the content
of the evaluation index system is not readily available, the relevant data are not easy to
be obtained, so the relevant principles should be followed in the study. In addition, the
index also should pay attention to the actual effect, and has certain guiding function to the
development of the tourism related policy. Article in constructing the evaluation index
system; mainly refer to the relevant experts and scholars research findings of tourism
strong province, tourism industry and Tourism Competitiveness. In the foundation of
the analysis of the factors affecting the construction of tourist province, based on the
existing statistical yearbook, tourism yearbook, social and economic yearbook, Tourism
Bureau data and other official authority data analysis and research. The main principles
of the study are:
1. Objectivity principle: In the selection of indicators and establishment of
mathematical model, as far as possible meet factor involved in the construction
of tourism province, and considering the accuracy indices and availability, based
on original data of statistical yearbook and Tourism Bureau, as far as possible
using data directly to avoid the subjective.
2. Systematic principle:Evaluation index system of tourism province to ensure
the logic and systematic, on the one hand, the index system can comprehensively
reflect the general characteristics of the strong province of the travel. At the
same time, it can reflect the inherent connection between the index, so that
the objectives and targets of evaluation can contact each other, constitute the
organic connection of overall.
3. Measurable principle:The strong province of the travel on the evaluation
indicators to have measurable and index system should as far as possible, clear
and concise, data can be easily obtained, calculation method of measure can
integrates the various indicators become a comprehensive index to simple and
clear, easy to manipulate and measure, data is accurate, easy to understand.
154 RISTI, N.º E5, 07/2016

System level A Element layer B Index layer C

C1Number of domestic tourists;C2 Number of
Number of tourist consumption B1
foreign tourists
C3 Domestic tourism income;C4 Foreign
Tourism consumption level B2
Tourism market tourism income
demand C5 Per capita consumption of tourists;C6
A1 Tourism consumption behavior B3
Average length of stay;C7 Revisit rate
C8Development potential of domestic tourism
Tourism consumption potential B4 market;C9Development potential of foreign
tourism market
C10 Quantity and characteristics of tourism
Tourism resource endowment B5 resources;C11 Popularity;C1Rare degree;C13
Grade degree;C14 Degree of development
C15 Product features;C16 product brand;C17
Tourism product development B6
Product attraction;C18 Additional product
Tourism supply
capacity C19 Total tourism income;C20 Tourism
Development potential of tourism
A2 enterprise tax;C21Total labor productivity of
enterprises
tourism enterprises;C22 Number of travel
B7
agencies;C23 Star Hotel quantity
C24 The quantity and quality of tourism colleges
Talent supply potential of tourism
and Universities;C25 Number of tourism
professionals B8
professionals
C26 Per capita GDP;C27 Tourism added
Macro-economic development value accounted for GDP;C28 Tourism
level Investment;C29Total consumption of social
B9 retail goods;C30 Engel coefficient of urban
residents
C31 Credit information publicity;C32 Balance
Tourism of financial revenue and expenditure;C33
Government attitude and
industry support Degree of perfection of legal policy;C34 Capital
management ability B10
force A3 investment;C35 Industry management level;C36
Publicity and promotion efforts
C37 students in all kinds of schools;C38 The per
capita funding for science and education;C39
Cultural development and scientific
Number of scientific and technological
and technological innovation B11
innovation in tourism industry;C40 The
construction level of tourism information system
C41 Traffic line density;C42 Toilet density in
construction of infrastructural tourist area;C43 Number of tourist facilities
facilities B12 owned by one thousand people;C44 Tourism
facilities density
tourism strong
C45 Location advantage;C46 Transportation
province security Location condition B13
condition
force
A4 C47Ecological environment carrying
capacity;C48 Tourism resources carrying
Environmental carrying capacity
capacity;C49Carrying capacity of community
B14
residents’ economic life;C50 Psychological
carrying capacity of community residents
RISTI, N.º E5, 07/2016 155

System level A Element layer B Index layer C

C51 Tourism revenue accounted for
GDP;C52Tourism food and beverage
Industry
contribution;C53 Tourism accommodation
contribution
contribution;C54 Contribution of tourism
Tourism pulling B15
and recreation;C55 Tourism shopping
effect A5 contribution;C56 Contribution
C57 Every ten thousand people in the number of
Employment multiplier
tourism related industries;C58 The number of
B16
jobs created by tourism
Table 7 – Evaluation index system of tourism strong province
4. Comparability principle: Tourism province construction evaluation

standard is evolving, the area due to inconsistent development. Its industrial
structure are not the same, in the selection index design need to consider factors
in time and space can be compared, data of different areas and in different
periods using the same standardized indicators were analyzed.
Subject starting from the elements of the strong province of the travel, the related
literatures at home and abroad were analysis and summary, based on that, preliminary
proposed 66 indexes evaluation of tourism strong province. To develop expert
consultation table, through the Delphi method to index were screened, and the correction
of some indicators, setting up a tourism province evaluation index system framework
by the tourism market demand and capacity, tourism supply capacity, tourism support
force, tourism security and tourism pulling effect of five layer system elements.
Acknowledgments
This paper is supported by Anhui Provincial Tourism Bureau key project “Study on the
indicator system of tourism”(SLYJKT201516);Humanities and social sciences research
key project of Anhui College “Evaluation and utilization of wetland tourism resources
in Northern Anhui”(SK2015A565); Anhui Province Department of education “teacher
studio”(2013jxms083).
References
Abraham, P., Natan, U. (2000). The intensity of tourist–host social relationship and its
effects on satisfaction and change of attitudes: the case of working tourists in Israel.
Tourism Management, 21(4), 395–406
Claudia, M. (2013). Prevention of travel-related food borne diseases: Microbiological
risk assessment of food handlers and ready-to-eat foods in northern Italy airport
restaurants. Food Control, 29(1), 202–207
156 RISTI, N.º E5, 07/2016

Hongjian, W. (2009). Research status of tourism strong province, theoretical basis and
evaluation system. Tourism Tribune, 5, 114–119.
Li, F. (2014). The construction of tourism strong province evaluation system and
empirical measurement research. Territory & Natural Resources Study, 1, 47–50.
Lijing, C. (2010). The path choice of tourism strong province construction. Dongbei
University of Finance and Economics, 11, 20–24.
Qiuyang, Y. (2009). Study on structural model and measurement of tourism industry
development potential. Journal of East China Normal University, 5, 114-119.
Informação, (10), 51-64.
Song, G., Liu, W. (2005). Economic analysis and test parameter system research about
optimization of industrial structure. Science Technology and industrial, 7, 6-9.
Xiangyang, F. (2009). Research on evaluation index system of regional tourism
competitiveness. Inner Mongolia science technology and economy, 6, 3-8.
Yongjun, T. (2008). Evaluate the strength of the tourism industry and the construction
of tourism strong province of Guangxi. Economic management, 2, 44-50.
Zhang, L. (2007). Optimization of industrial structure and evaluation of quantitative
research methods about our country. Shandong University of Science and
Technology, 1, 62-65.
Zhang, H. (2014). Rural tourism in shandong province SWOT analysis and
countermeasures research. Market modernization, 24(5), 30-34
Zhifa, W. (2011). The connotation and Industry characteristics of tourism strong
province. China Travel News, 7, 10-13.
RISTI, N.º E5, 07/2016 157

Research on Sports Tourism Resources Value and

Promotion with Skating and Skiing based on Data
Mining Method
Xianfeng Zha 1, Xiaomei Zhang 2,*
* 915428334@qq.com
1
Department of Physical Education, Northeast Agricultural University, Harbin 150030, China
2
Pages: 158–170
Abstract: With the development of economy and the improvement of living

standards, the electronic platform provides more opportunities for tourism
industry. The tourists are change from a single sightseeing tour to all kinds of special
tourism such as sports tourism. In this paper, we analyze the ice-snow tourism
especially in Heilongjiang province, as it has climatic conditions and unique ice
and snow tourism resources. After analyzing the characteristics of ice and snow
tourist resources, we build the evaluation index system of tourism resources value.
By using factor analysis, we find out that the elements of ice and snow tourism
resources value mainly includes economic factors, cultural factors and ecological
factors. On this basis, we put forward relevant policy suggestions.
Keywords: Data mining, Sports tourism, Resources value, Skating, Online
survey, Ice-snow tourism industry
1. Introduction
Since the reform and opening up, China’s tourism industry has achieved great
development. From 1978 to 2015, China’s tourism market has expanded the scale of
66 times, the annual growth rate reached 16.8%, tourism foreign exchange income
average annual growth of 19.1%. On the other hand, with the increase of economy and
the improvement of living standards, people’s leisure time in recent years, especially in
the Spring Festival, during the golden week, domestic residents showed a strong tourism
demand. The World Tourism Organization predicts that China is expected to in 2020
to become first world tourist reception country and the fourth largest tourist country
abroad. In this context, one of the main trends in the development of tourism, tourists
are from single sightseeing tourism to the holidays and all kinds of special tourism such
as sports tourism transformation (Urioste, 2015; Lopes, 2015). Especially the 2001
World Tourism Group, the fabric and the International Olympic Committee held a world
of sports and tourism conference, sports tourism has become another focus of ecological
tourism circle. The ice tourism is an important part of sports tourism, which is one of
the most rapid development projects. Taking Heilongjiang Province as an example, the
158 RISTI, N.º E5, 07/2016

province due to the special geographical environment, climatic conditions and history,
have unique ice and snow tourism resources. In recent years, the development of
tourism industry in Heilongjiang province is very rapid; the provincial government put
the development of tourism as a pillar industry of the national economy, more quickly
transformed into economic advantages of tourism resources.
Figure 1 – Ice and snow tourism in Heilongjiang

At the same time, with the expansion of ice and snow tourism scale, contradictions
and problems in all aspects and levels of restricting the development of the tourism
industry to further highlight the snow (Tapsuwan, 2015; Polat, 2015; Wyss, 2014).
From a practical point of view, the ice and snow tourism in the expansion of the existing
tourism resources development at the same time, disorderly and wasted, predatory
development and operation, the serious phenomenon of repetition development
and economic benefits of ice and snow tourism such as poor. Especially the ice snow
tourism resource rich areas are located in remote, in addition the ice and snow tourism
resources development planning unscientific and unreasonable evaluation, affected the
further development of ice and snow tourism. From the theoretical point of view, the
reasons for these problems are that of ice and snow tourism resources and its law is not
clearly understood(Li, 2014; Siow, 2014). First, ice snow tourism resources relatively
vague concept, lack of unified and acknowledged definition, no consistent classification
standard and scientific classification system, lead to the ice snow tourism resources and
its value do not have a clear understanding, is not conducive to scientific and reasonable
development. The ice snow tourism resources value also awareness of the existence of
bias (González,2015; Kariminia,2013). At present, the main discussed is economic, and
ignored the tourism is also a kind of cultural activities, the essence is aesthetic pleasure,
is the tourists as the main body of tourism with the help of tourism media and other
external conditions, through the activities of tourism object dynamic, to realize their
own a need for non-domiciled in a dynamic process of complex. Therefore, the ice and
snow tourism culture is showing its importance in ice and snow tourism development
status. Secondly, no clear understanding of the mechanism on the formation of ice and
RISTI, N.º E5, 07/2016 159

Research on Sports Tourism Resources Value and Promotion with Skating and Skiing based on Data Mining Method
snow tourism resources value(Liggett, 2011;Healy, 2016; Varley, 2011)). Only think of
snow and ice is the view of the nature of the product makes the understanding of the
value of the snow and ice resources stay on the surface level, the lack of other prompted
snow and ice have become with analysis of the mechanism of the value of resources,
resulting in the lack of development of theoretical research on the special resource,
which has an effect on the realization of the ice snow tourism resources value. What are
the factors to the snow and ice resources contain potential can be reflected, to realize
its value is how to effect. These problems are related to whether in the development
process of ice and snow tourism resources, bring economic, social and cultural benefits
of important problems scientifically and reasonably, the overall planning. In short, the
lack of scientific theoretical system of a complete set of ice and snow tourism resources
value, it is necessary for further discuss.
2. Data mining and neural network

2.1. Neural network
Connectionism school neural network belongs to the field of artificial intelligence, is an
application similar to the brain synaptic structure information processing model, data
mining process as shown in figure 2. Classification or regression models are constructed
in the form of tree structure. It divides the data set into smaller subsets, and the related
decision tree is gradually developed. The end result is a tree consisting of a decision
node and a leaf node. A decision node has two or more branches. Leaf nodes represent
a classification or decision. The top of the decision node is the best predictor of a root
node. Decision trees can handle categorical variables and numeric variables, the decision
tree as shown in Figure 3.
160 RISTI, N.º E5, 07/2016

Figure 3 – The decision tree
2.2. BPTT algorithm

Neural network computing power or “storage knowledge” is reflected in the connection
weights between neurons. Therefore, learning algorithm is the method of regulating the
network weight, that is:
wnew = wold + ∆w (1)
The computational tasks of the network are usually quantified as the optimization of
some performance function J (W), therefore:
wnew = wold + α ⋅∇ w j (2)

k output and target output error can expressed as:
d ( t ) − yk ( t ) k ∈ T ( t ) (3)
ek ( t ) =  k
 0, otherwise
The time of the t can be expressed as:
2
1 n
J (t ) = ∑ ek ( t )  (4)
2 k =1 
The total performance function of the network is:
t1

J ( t o , t1 ) = ∑ J (τ )
τ = t0 + 1
(5)
Process of Epoch-wise BPTT algorithm is as follows:
∂J ( t0 ,t1 ) t1
∂J ( t0 ,t1 )
∂wij
= ∑
τ = t0 + 1 ∂wij (τ )
(6)
RISTI, N.º E5, 07/2016 161

Conclusion using the chain rule:
∂J ( t0 ,t1 ) ∂J ( t0 ,t1 ) ∂sk (τ ) (7)

∂wij
= ∑
k∈U ∂sk (τ )
×
∂wij (τ )

Error as:


f′ ( s (τ )) e (τ ) ,τ = t
k k 1
δ k (τ ) =   n  (8)
(
 f ′ sk (τ ) )∑  wkj δ j (τ + 1 ) + ek (τ ) 

  j =1 
The gradient of the network performance function in the weight space is:
t1
∇ w J ( t 0 , t1 ) = ∑ δ (τ ) y (τ − 1 )
τ = t0 + 1
i i
(9)
3. Classification and characteristics of ice and snow tourism resources

3.1. Characteristics of ice and snow tourism resources
The value of all the resources itself has many functions; the added value of tourism
value is just one of the ancillary uses. The first is for minority people to pursue a
comfortable choice embodied, and then again with enhanced the scarcity of the
resources, the resources of additional value so as to continuously improve and be more
people know and choose. The selection process of the resource value, also determines
the creation and Realization of value of tourism resources. The value of tourism
resources is to meet the initial expression of a people in the process of viewing, travel
and entertainment and get to know the exotic. From an economic perspective, reflect
the value of tourism resources is more essential in a willingness to pay people for
environmental comfort, visual enjoyment or culture of infection.
From the perspective of the classification of ice and snow tourism resources, including
the natural tourism resources and cultural tourism resources, so it has the characters
of natural tourism resources and cultural tourism resources value. Compared with the
general commodity value, has its own characteristics and value of tourism resources:
1. Scarcity: This scarcity from most of the ice and snow tourism resources is
limited, the cost of human development and protection of their human needs
and constantly increasing endless.
2. Overall usefulness: Some features of ice and snow tourism resources as a
whole may play a role, not a change in composition of snow resources in the
system while the other factors remain unchanged.
3. Sharing: Sharing for tourists to enjoy the ice and snow tourism resources
formed by the landscape, this property makes people in the ice and snow
tourism, it cannot exclude others to enjoy sightseeing.
4. External: Including the external economy and external economy. External
economy shows that the tourism activities to others and the society benefit,
162 RISTI, N.º E5, 07/2016

while the beneficiaries do not need to pay for this, such as the promotion of
the development of the economy, driven by the improvement of infrastructure,
improve environmental quality, enhance the visibility and so on. External
diseconomy shows that tourism activities of others or social losses caused by
external diseconomies of people without paying cost. If the conflict caused by
the culture, bring inconvenience to people’s life and environment pollution.
Figure 4 – Characteristics of ice and snow tourism
3.2. Classification of ice and snow tourism resources

From the view of connotation, ice and snow tourism resources mainly include two
categories: (1) natural tourism resources: formed in the climate, geomorphology and
hydrology and other natural factors under natural resources such as snow, ice, and
other natural resources and the combination of mountain and forest caused by the
ornamental, including the snow and rime; (2) cultural tourism resources: Based on
natural tourism resources, after processing, and combined with the specific regional
culture, customs, arts and the formation of tourist attractions, including a variety of
winter sports and recreational activities as well as the characteristics of the customs,
such as ice, snow, snow, ski skating, ice and snow tourism festival activities, etc.. It can
be concluded that the snow tourism resources to tourism based on ice and climate and
the main attraction of a special form of tourism, the basic types include ice and snow
activities of two kinds of landscape. Ice snow landscape natural landscape such as snow
village, tree and snow and other artificial landscape such as ice, snow and ice snow
sports activities including entertainment. The ice and snow entertainment including
skating skiing, sailing, ice and snow sports including hockey, winter swimming, skating,
hunting in the winter, skiing and other. In addition, the ice carried out based on the
different forms of economic activities, attract many people to participate, not only to
promote economic development, and promote the development of the third industry,
the objective is also part of the ice and snow tourism resources. Such as the use of ice
and snow tourism area personnel intensive advantages, hold various forms of trade
RISTI, N.º E5, 07/2016 163

fairs and expositions, etc.. But we should also see, it is owing to the differences in the
nature tourism resources of diversity and different resources between the existences,
any classification will not be exhaustive or covers all types of resources. In addition,
the non-uniqueness of classification principle, determine the classification results
between different and incompatible cross. Therefore, the classification of ice and snow
tourism resources should not be advised to Jane fan, wide should not narrow, should
develop and utilize directly reflects the value of resources.
Figure 5 – Cultural tourism and snow tourism

The traditional definition of natural tourism resources and cultural tourism resources is
its connotation and core. With the development of the tourism industry to the depth, the
connotation of tourism resources will be more and broader; the type will be more and
more. On this basis, the classification system of CSSR ice and snow tourism resources
has been developed, including:
1. Core resources: that is constitute the core of ice snow tourism resources,
those can be developed to be meet ice and snow tourist specific or comprehensive
traveling motivation, tourism resources of the ice and snow tourist core
interests need to type products. Core tourism resources elements: the snow
and ice resources of high grade (usually special grade), distinctive theme, rich
connotation, long life cycle, transportation accessibility is good, tourists carrying
large amount.
2. Supportive resources: refers to the formation of ice snow tourism more
direct support resources, to directly support the ice snow tourism to mold, the
ice snow tourism line construction, rich ice and snow tourist visit content and
function, strengthen the ice and snow Tour theme, providing a more diverse
service resources.
3. Subsidiary: resources: it is a supporting snow tourism resources of the
indirect support resources belong to the ice and snow tourism. Although in
tourism to more peripheral regions, but in the corridor of tourist traffic routes, to
core resource nodes provide meet the need and multiple interests and the depth
164 RISTI, N.º E5, 07/2016

of experience, the increase in ice and snow travel sojourn time, to construct the
ice snow tourism broadband, spheres or network system resources.
4. Reserved resources: refers to the formation of ice snow tourism to reserve
resources, have a certain grade, value of development and utilization, but because
in the same or similar core resources, alternative energy sources, resources and
seize the opportunity of the shadow covered resources. These resources are
increased by the presence of ice and snow tourism resources abundance, the
market capacity and scale, can be reference for all kinds of resources, alternative
funds in the future.
4.1. Elements
Ice and snow tourism resources can be divided into natural tourism resources and
cultural tourism resources, so it has the general nature in addition to the value of
tourism resources, has its unique value. In order to determine the elements of the ice
snow tourism resources value, this paper follows the some indexes of natural tourism
resource value evaluation both at home and abroad, such as size and abundance.
According to the characteristics of regional, seasonal ice and snow tourism resources
outstanding, set the appropriate travel range and appropriate travel cycle variables. In
addition, the ice snow tourism resources as a kind of humanities tourism resources, a
strong cultural connotation, including the contents of the snow art, architectural culture
of ice and snow, ice and snow dress culture, diet culture of ice and snow, ice and snow
entertainment culture and ice science and technology and culture, an important level
is the art of ice and snow from the form, source, such as in the fifteenth day lantern
show ice art. Then we extract snow landscape design art of indicator variables. The
ice snow tourism facilities, tourism resources, tourism products and ice snow tourism
managers in the management process reflected the cultural connotation, called snow
tourism object culture. We make evaluation on the ice and snow tourism object culture
from the following several indicators to measure, as shown in table 1. A complete
evaluation index system can reflect the ice and snow tourism resources value aspects. As
to elements of ice and snow tourism resources value recognition accurate, here follow
as comprehensive as possible to the principle of constructing evaluation index system.
Index Variable Index Variable

Landscape peculiar X1 project design X9
Ecology X2 suitability X10
Artistry X3 travel cycle X11
Recreational degree X4 travel range X12
Abundance X5 Service quality X13
Visibility X6 integrity X14
heterogeneity X7 Environment X15
theme X8 coordination X16
Table 1 – The evaluation index system of ice and snow tourism
RISTI, N.º E5, 07/2016 165

4.2. Factor analysis

According to the framework of the index system of the construction designs the
questionnaire, were selected and the Heilongjiang Province four ice snow tourism to
a total of 400 visitors a survey and interview, investigating the use of Likert 5 scale
table, please be investigated according to their own experience on the ice snow tourism
resources value for each index score, for each indicator from the “very low” to “very high”
degree, set 1-5 a total of five digital said, one that is very low, 5 indicates a very high. The
survey data obtained by the factor analysis method for processing, the factor analysis is
some perplexing relationship variables attributed to several comprehensive factors, in a
multivariate statistical relationship between the reproduction of the original variable and
factor analysis method. The factor analysis method can classify the variables according
to different factors, a statistical method which belongs to multivariate analysis process
of dimension reduction. The basic idea is through the study of the internal structure
variable correlation coefficient matrix; find the control variables a few random variables
to describe the relationship between several variables.
Variables Tourist Ⅰ Tourist Ⅱ Tourist Ⅲ Tourist Ⅳ
X1 0.253 0.136 0.154 0.171
X2 0.352 0.229 0.254 1.014
X3 0.025 0.586 0.046 -0.344
X4 0.032 0.241 0.092 -0.326
X5 0.855 -0.105 0.445 -0.072
X6 -0.742 -0.079 0.245 0.943
X7 0.265 0.163 0.177 -0.434
X8 0.491 0.729 0.469 0.286
X9 1.396 0.581 0.461 0.344
X10 1.472 0.549 0.069 0.362
X11 1.032 -0.075 0.138 0.235
X12 0.566 0.079 -0.046 0.217
X13 0.113 -0.037 -0.069 -0.325
X14 1.799 0.597 0.161 0.109
X15 -0.289 0.229 0.223 0.642
X16 -0.327 -0.628 0.107 0.199
Table 2 – Standard data evaluation of tourism resources
166 RISTI, N.º E5, 07/2016

Serial Characteristic value Variance contribution Cumulative contribution

rate% rate%
1 8.532 49.561 49.561
2 3.386 22.045 71.606
3 1.561 20.192 91.798
4 0.704 2.286 94.084
5 0.682 1.817 95.901
6 0.453 1.152 97.053
… … … …
16 0.015 0.423 100.000
Table 3 – Eigenvalue and cumulative contribution rate

Through factor analysis shows that the three major factors of evaluating the value of
resource of travel of ice and snow, which factor 1 has seven indexes higher load, the
seven indexes in turn is unique landscape level, suitable for recreation degree, size
and abundance, visibility and influence, the appropriate travel period, the appropriate
travel scope, quality of service, these indicators in largely reflects the value of ice snow
tourism resources in the economy, so called for economic factors. Factor 2 in the
five indexes higher load, the five indexes in turn is landscape design art, tourism and
cultural heterogeneity, the theme of cultural conception, project design in culture, show
the suitability, these indicators reflect the is value of resource of travel of ice and snow
culture, so called for cultural elements. Factor 3 has four indexes higher load, the four
indexes in turn is landscape ecology, ecological resources of integrity, environmental
protection, ecological environment of coordination, and these indicators reflect is ice
snow tourism resources and ecological value, so called as ecological factors.
Factors High load indicators Named

Factor 1 X1、X4、X5、X6、X11、X12、X13 economic factors
Factor 2 X3、X7、X8、X9、X10 cultural elements
Factor 3 X2、X14、X15、X16 ecological factors
Table 4 – Load factor classification

Through the above analysis of cultural elements are important elements of ice and snow
tourism resources value, show that the ice and snow tourism resources value has a strong
cultural connotation. This conclusion is not explicitly put forward the former research
institute. Economic factors mainly refer to as exotic landscape level, suitable for recreation,
appropriate travel period reflect elements of ice snow tourism resource economic value;
cultural elements mainly refers to like the concept of landscape design art, tourism and
cultural heterogeneity, the theme of cultural reflect ice snow tourism resources value and
culture value elements; ecological factor mainly refers to as landscape ecology, ecological
resources of integrity, environmental protection etc. reflect ice snow tourism resources
ecological value elements.
RISTI, N.º E5, 07/2016 167

5. Conclusions
The ice and snow tourism resources as the research object, to explore the formation,
implementation and promotion of ice and snow tourism resources value from
theoretical and empirical perspective. To identify the elements of ice and snow tourism
resources value, establish the system of ice and snow tourism resources value. Based on
the analysis of ice and snow tourism resources and the value concept, through factor
analysis, pointed out that the elements of ice and snow tourism resources value including
economic factors, ecological and cultural elements. Accordingly, we make the analysis
system of ice and snow tourism resources value from economic value, ecological value
and cultural value, which belongs to the practical value of economic value, ecological
value and cultural value of the potential value. To construct the system of ice and snow
tourism resources value, clarifies the structure of the ice and snow tourism resources
value, to further lay a good foundation for further analysis. Especially the cultural value
into the value system highlights the particularity of ice and snow tourism resources
value, which is of great significance for expanding the meaning of value.
5.1. Enhance the Ice and snow tourist culture connotation

Ice and snow tourism national characteristics and regional characteristics of
the special cultural background of ice and snow tourism, which once formed it has
relative independence and stability, has become the main motive of tourist experience of
snow tourism. The shape of ice and snow tourism image connotation is to create a culture
of ice and snow tourism culture, national characteristics and local characteristics of
tourism special cultural theme. Therefore, whether it is the natural landscape and human
landscape should have its definite theme or by refining the design theme, the theme will
be after the particular concept has become the soul of tourism. Theme tourism culture
has laid the foundation for the building of tourism resources development and tourism
culture. In turn, by designing the shape of the tourism image, can shift the cultural
theme and connotation of tourism destination for tourists cognitive image. Specifically,
the theme positioning of the tourist destination is through tourism and culture to fully
evaluate the tourist to find local tourism culture and culture tourism other differences,
based on the local culture of tourism. And, because of the culture in the development of
tourism, cultural tourism is not immutable and frozen, need to adhere to the innovation
of tourism culture in the theme positioning.
5.2. Improving the ice-snow tourism industry system

In the Heilongjiang ice snow tourism resources development process to emphasize the
standardization, vigorously regulate the market, establish good word of mouth publicity
and encourage tourists to travel again and drive the surrounding people to travel here
and the project has been developed. In addition to the development of a variety of snow
tourism products, tourism reception quality is another important factor. The level of the
destination reception includes tourist service facilities and service quality. The tourists
from ice and snow tourism market in Heilongjiang Province mainly from Hong Kong,
Guangzhou, Beijing, Shanghai and other places. These places are the economically
developed areas, the service facilities and service level are higher than the province,
accustomed to the service quality of the tourist reception facilities and service quality
168 RISTI, N.º E5, 07/2016

is very sensitive. For example, the tableware cleaning, cooking dishes degree greasy,
hotel shower facilities, public toilet hygiene, are very sensitive. Therefore, in order to
develop tourism market, enterprise should make particular attention to the quality of
the reception facilities.
5.3. Strengthen the government’s macro-control

Because of geographical reasons, the Heilongjiang provincial government in the ice
snow tourism resources development process occupies very important role, with the
development of market economy, the government needs to change function, further
strengthen macro-control and management in the industry, to focus on strengthening
the development of ice and snow tourism resources of macro planning, snow tourism
market management norms, eliminate redundant construction and vicious competition
phenomenon. Because the ice and snow tourism activity is very challenging and
dangerous, so must keep in mind the concept of safety first. This requires strengthening
the government supervision mechanism, develop hard and soft conditions of ice and snow
tourism business activities of enterprises, and in strict accordance with the standards of
examination and approval. Long term then tourism enterprise must consider from the
development of the tourism industry, attaches great importance to the safety problem
of ice and snow tourism activities, to ensure that the property and personal safety of
tourists. In addition, a sound legal system of tourism in Heilongjiang province is ice and
snow tourism development need to solve the problem as soon as possible.
5.4. Develop regional tourism projects

With the improvement of people’s living standards, urban residents changed the
holidays; travel has become a new choice of many families, however, long distance
tourism by the transportation, accommodation, time and other factors limit, short ice
snow tourism of urban residents is more attractive to potential. Liaoning Province,
Jilin province and Heilongjiang province some city population is numerous, has a great
demand for ice and snow tourism. In addition, in all kinds of schools during the summer
vacation, the students have to choose the appropriate venue for ski snow sports, enrich
the basic requirements of life. So according to market levels, the development of regional
tourism project has to meet local tourists. Heilongjiang Province in the development of
ice and snow, there are two main categories of resources, in view of the snow resources
development with open space, large environmental capacity, participation, the thrills
of, can arouse the broad masses of ice and snow tourist interest, and therefore should
be to focus on the snow, can proper consideration add some strong participation, low
price of tourism products, to meet the Heilongjiang Province of ice snow tourism of
increasing demand.
Acknowledgments
This paper is supported by Heilongjiang Province philosophy social science planning
projects: “The study of folk sports into the physical education curriculum in rural schools
in Heilongjiang province” (15TYE01). Heilongjiang higher education reform project:
“the construction of teaching model of college sports club”, (JG2014011190).
RISTI, N.º E5, 07/2016 169

References
González, M., González, L. (2015). The co-creation as a strategy to address IT governance
in an organization. RISTI-Revista Ibérica de Sistemas e Tecnologias de Informação,
(14), 1–15.
Healy, N. (2016). Low versus high intensity approaches to interpretive tourism planning:
The case of the Cliffs of Moher, Ireland. Tourism Management, 52, 574–583.
Kariminia, S. (2013). Environmental Consequences of Antarctic Tourism from a Global
Perspective. Procedia - Social and Behavioral Sciences, 105, 781–791.
Li, G., Yang, X. (2014). Destination island effects: A theoretical framework for
the environmental impact assessment of human tourism activities. Tourism
Management Perspectives, 10, 11–18.
Liggett, D. (2011). From frozen continent to tourism hotspot? Five decades of Antarctic
tourism development and management, and a glimpse into the future. Tourism
Lopes, P., Pacheco, S. (2015). Fisheries, tourism, and marine protected areas: Conflicting
or synergistic interactions?” Ecosystem Services, 16, 333–340.
Polat, N. (2015). Technical Innovations in Cruise Tourism and Results of Sustainability.
Procedia-Social and Behavioral Sciences, 195, 438–445.
Siow, T., Luh, D. (2014). A taxonomy of creative tourists in creative tourism. Tourism
Tapsuwan, S., Rongrongmuang, W. (2015). Climate change perception of the dive
tourism industry in Koh Tao island, Thailand. Journal of Outdoor Recreation and
Tourism, 11, 58–63.
Urioste, S. M., Scaccia, D. (2015). Exploring visitor perceptions of the influence of
climate change on tourism at Acadia National Park. Maine, Journal of Outdoor
Recreation and Tourism, 11, 34–43.
Varley, P. (2011). Ecosophy and tourism: Rethinking a mountain resort. Tourism
Wyss, R., Abegg, B. (2014). Perceptions of climate change in a tourism governance
context. Tourism Management Perspectives, 11, 69–76.
170 RISTI, N.º E5, 07/2016

Design of Forest Tourism Resources Evaluation

System based on GIS
Chuanming Yang, Xiaomei Zhang*
* 915428334@qq.com
Northeast Agricultural University, Harbin 150030, Heilongjiang, China

Pages: 171–181
Abstract: Geographic information system (GIS) platform can make the evaluation
of forest resources combined with geographic information, and some indexes of
forest resources evaluation can be obtained through the spatial analysis and query
function of GIS. And at present, the forest tourism industry as an important part
of the green industry, forest leisure tourism has gradually become the of China’s
tourism market. Through quantitative evaluation, the results show that the weight
of resources value is 33.4, the main factors are ornamental value and experience
value. Tourists choose the category of tourist resources in order to take a vacation
and eco-tourism, travel time is mostly selected in the individual off holiday, and half
of the tourists are willing to travel with friends and relatives.
Keywords: Geographic information system, spatial data, forest tourism, leisure
vacation
1. Introduction
Forest tourism has been more and more people attention. At the same time, productivity
has been great development, people’s living standards have markedly improved,
the increase in disposable income and leisure time increasing, the concept of leisure
gradually formed (Mattheus,2013; Petr,2015; González, 2015). The leisure tourism is
being accepted by more and more people. Leisure tourism has gradually become the main
trend of the development of China’s tourism industry. Data show that in the developed
countries of the world tourism, leisure tourism is the main body of the tourism market,
accounting for 50% of the total market (Kostas, 2013). At present, an important content
of modern forestry and forest tourism industry as a green industry is an important part
of the essential, leisure tourism in China accounted for only about 20% of the overall
market, forest leisure tourism has gradually become an important direction of the
mainstream and tourism to the tourism consumption in China. Therefore, it is of great
significance to study the development of forest leisure tourism products for our country
to develop the leisure tourism market and realize the goal of the world tourism power
(Sandro, 2013; Shihong, 2015; Ashutosh, 2015). Forest tourism product is the subject of
forest tourism enterprises of all business activities, deeply and systematically research
and development of its, is not only the forest tourism marketing of the starting point
and basis for formulating relevant strategies, but also to ensure that the competition
RISTI, N.º E5, 07/2016 171

Design of Forest Tourism Resources Evaluation System based on GIS
invincible important guarantee of enterprise in the fierce competition in the market

(Romain, 2015; Mukesh, 2015). This study reviewed the connotation, development and
products of forest tourism, leisure tourism, forest leisure tourism, and so on.
Figure 1 – Forest leisure tourism

In the forest resources assessment work, the general means of using manual methods
for data collection and analysis. Because of the particularity of forest resources,
the more difficult to implement, requires a lot of manpower and resources, and the
calculation result of long cycle, the evaluation results lag relatively strong. At the same
time, most of the evaluation index data need to go through multiple steps of statistical
calculation; manual computation also has poor precision and low efficiency. Moreover,
the evaluation of forest resources manual implementation results are only abstract
figures could not be associated effectively with the actual geographic information, there
is no evaluation of spatial information support of the difficult for the forest resources
management decision-making support with the development of the country. “Digital
forestry” project as well as the technology matures, the forestry informationization has
become an inevitable trend. In recent years, although the For the evaluation of forest
resources information has been studied and achieved some results, but the combination
of GIS and is relatively small. And the GIS platform support can make the evaluation of
forest resources combined with the geographical information display, more intuitive,
and the evaluation of forest resources through some indicators of spatial analysis of
GIS query to obtain the function. Therefore, based on the GIS evaluation system of
forest resource construction for real work is particularly important. The forest resource
evaluation system based on the data of forest resources survey, therefore, the basic unit
classes as the forest resources assessment. According to the demand, the system adopt
value evaluation method, the full range of data value of wood charcoal the output value
of forest products, the output value, the value of water conservation, soil conservation
value, forest carbon fixation and oxygen release value, air purification value evaluation
etc. At the same time, in order to improve the accuracy and efficiency of evaluation, to
provide support for forest management and decision-making.
172 RISTI, N.º E5, 07/2016

2. Spatial analysis of Geographic Information System (GIS)

2.1. Geographic Information System
ARC/INFO is developed by the United States environmental system research, at present,
the world is widely used to promote the use of more typical general GIS software,
has been more than 30 years of history, applied to more than 100 areas. Arc / info is
described by the feature map and the topological relations of arc system and recording
properties data of relational database management system NIFO combined with “such a
mixed data model taking into account the characteristics of spatial data and non-spatial
data, the two different kinds of data, effectively realized the operation of two different
kinds of data, processing and management. In the structure of the software using the
basic module of the toolbox to achieve the input! Analysis, management and display
output functions, is an ideal tool for GIS. W is the United States Environmental Systems
Research Institute in recent years to meet the new computer technology development of
new products, can be used for personal computers and computer workstations, but the
majority of personal computer users ARCV. It uses the window menu system, the user
does not have to remember thousands of complex commands, and therefore is a user
friendly system. It can be directly read into the data generated by ARC/NIFO, so the two
can interact with the use of. Data input, editing, topological relations and spatial overlay
of raster data, digital elevation model (DEM) are completed in ARC/NIFO, the other is
carried out in ARCVEIW.
Figure 2 – Geographic Information System
2.2. Layer management

Layer is entity files of the GIS spatial data, from the perspective of graphic management,
according to the characteristics of forestry will have the same feature primitives are
put together to form layers. The layer is strictly divided into three types: point layer,
line layer, and surface layer. Project is a collection layer file, it contains the name of
each layer, layer properties and layer state, save the drawing environment of all kinds
of information such as symbol library, color library, help management in a complete
mapping of the shell, the project file through often does not directly manage the spatial
RISTI, N.º E5, 07/2016 173

data. To the forest resources spatial database based, using the map editor and secondary
development tools, form a thematic map of forest resource, forest terrain forest map.
Figure 3 – GIS simulate landscape
Figure 4 – GIS software
3. Forest leisure tourism

3.1. General categories
Tourism products, which are based on the development of natural tourism resources,
which are based on the forest tourism resources or the forests, are called the forest
tourism products. The types of forest tourism products are rich and varied; focusing
on the participation of people involved in the product can reflect the function of forest
leisure tourism products. These include forest leisure, forest recreation, forest health and
leisure, forest recreation and leisure, forest adventure and leisure and forest ecological
education and leisure.
174 RISTI, N.º E5, 07/2016

Level one Level two content Material carrier

Natural tourist Primitive forest landscape, geological Natural landscape
Sightseeing type landscape, fountains, waterfalls, etc. resources
tourism Enhance tourism Botanical garden, wild animal park, bird Animal and plant
type garden, butterfly resources
Tent Inn, Log
Country play Cabins, tent, camping, picnic and barbecue
House Hotel
Skiing, skiing,
Exercise fitness Water sports, mountain sports
hunting field
Forests,
Health care and
Forest bath, mountain convalescence hot springs,
recuperation
Activity mountains
participation recover one
tourism ‘s original Hiking, fruit picking, farming Forest, wilderness
simplicity
Adventure,
Adventure Cave exploration, speed boats, high-altitude
mountain, virgin
stimulus landing
forest
Popular Science Production of specimens, animal and plant Botanical garden,
Education knowledge botanical garden
Table 1 – The ordinary type of forest leisure tourism products
3.2. Characteristics of forest leisure tourism products

•• Comfort: forest leisure tourism should be a real relaxation, entertainment and
enjoyment of the process, is a special kind of tourism experience. Forest leisure
tourism products of comfort are different from other tourism products, more
emphasis on tourists psychological relaxes and physical health, so that tourists
can not only enjoy the comfortable environment and tourism reception facilities,
can enjoy quality services and facilities of the leisure entertainment facilities,
and rich activities. Experience a relaxed, relaxed atmosphere, to achieve a
complete relaxation of the body and mind. Forest leisure tourism products of
the comfort requirements of forest tourism facilities more complete more rich,
more quality services.
Figure 5 – Tents, camping
RISTI, N.º E5, 07/2016 175

•• Participation: forest leisure tourism is a special kind of tourism experience,

and it is an important way for tourists to get the experience. Through a fancy
sightseeing tour to make tourist get real tourism experience, tourists only to take
part in the forest tourism activities, such as sports fitness, sports exploration to
merge with the casual atmosphere, a pleasurable experience. The psychological
demands of the participatory forest recreational tourism product requirements
of forest recreation activities and visit the design consideration of tourism can
form the interaction between tourists and tour operators, tourists and tourism.
Figure 6 – Forest sports adventure
•• Selective: Forest leisure tourism has certain randomness, tourists prefer to

arrange trips and activities, for more and more fit, this is particularly important.
Forest tourism must be carried out according to the route, but the forest leisure
tourists can arrange themselves, participate in which leisure and entertainment
activities, choose which reception facilities, etc.. The daily activities arranged
by the tourism managers should be just a set of recommendations, and the
forest leisure tourists can design a plan according to their own preferences.
The selectivity of forest leisure tourism products require tour operators to
provide facilities, services and activities project varieties more abundant types;
requires visitors to tourism in the forest of time is more abundant, flexible travel
arrangements for the trip room.
•• Periodicity: Compared to periodic and rely on manpower development of
tourism products, forest tourism product generally has a longer cycle of,
participate in the experience tourism products is sightseeing type products have
a longer life cycle, and in tourism, nature tourism product life cycle and longer
than the man-made landscape tourism products.
•• Seasonal: Because forest tourism mostly occurs in the field, the impact of
climate change on forest tourism products show a strong seasonal. Due to the
non storage of products, seasonal severe impact on the utilization of resources
and the annual production potential of the product, so seasonal is an important
factor affecting the development of products. So that the forest tourism products
can be divided into seasonal reasons: the forest tourism resources which
constitute the products change with the change of climate, the seasonal. Such as
176 RISTI, N.º E5, 07/2016

flowers, forest bath, ski. One is the product depends on the forest environment
with climate change and changes in product performance of the seasonal. Such
as climate change caused by changes in water temperature and drifting cannot
be consumed in the colder weather.
Figure 7 – Waterfall view
•• Forest leisure tourism products of tourist attraction survey shows, fashion

leisure tourism projects to tourist attraction is the largest, accounting for 49.1%,
mountain climbing and other sports and leisure projects is more popular in the
leisure projects and other leisure and tourism projects accounted for roughly
the same; in the preferences of the survey of shopping and leisure products,
visitors of tourist souvenirs of process love, travel food and local specialties.
Shopping
Tourist resources Ratio (%) Leisure travel Ratio (%) Ratio (%)
products
Fresh air and forest Mountaineering,
27.3 31.6 Tourism food 36.7
landscape forest exploration
Chinese herbal
Landscape scenery 23.6 Boating, fishing 22.4 21.0
medicines
Camping, farm
Drift, rock climbing 49.1 24.3 Keepsake 42.3
farming
Table 2 – The attractive investigation of forest leisure tourism products
4. Quantitative evaluation
4.1. Index weight
Quantitative evaluation of the intrinsic quality of all kinds of independent sites,
landscape entity, evaluates the expert evaluation method of analytic hierarchy process
(AHP). Only in this way can the complex of tourism resources is decomposed into
several levels and layer by layer analysis, one by one compared to establish a tree
type structure, obtained the weight value and evaluation, the subjective judgment in
quantitative expression and processing can be, is a combination of quantitative and
qualitative methods. In which the weight value is determined by the average weight
RISTI, N.º E5, 07/2016 177

method, that is, the evaluation of the comprehensive evaluation of the value of the
resources are 0.33, with a total score of 100 to give the weight of each factor scores, the
quantitative evaluation of ecological tourism resources in table 3.
Comprehensive Highest
Evaluation project Score Evaluation factor Score
evaluation score
Ornamental value 11.1 Pleasant sensation 3.7
novelty 3.7
Experience value 11.1 Exercise fitness 2.8
Adventure stimulus 2.8
Resource value 33.4
Country play 2.8
Popular Science
11.2 Cultural inside story 5.6
Education
Scientific value 5.6
Scale of scenic spot 6.6
Biological species 6.7
Tourism
33.3 forest coverage 6.7
environment
Tourism comfortable
6.6
period
Traffic communication 8.3 convenient 4.2
security 4.1
Tourism condition 33.3 Board and lodging
8.3
condition
Travel expenses 8.3
Table 3 – The attractive investigation

According to the parameter list to each factor classification and grading, and gives the
score standard, this paper uses the fuzzy score method to comprehensive evaluation
of a scenic spot, landscape entity score calculation model using the following
mathematical model:
n
E= ∑Q P
i =1
i i
In the above formula, E is the comprehensive evaluation value of tourism resources:

Qi is I evaluation factor weight; Pi is the first I evaluation factor; n is the number of
evaluation factors.
4.2. Factor analysis

Forest leisure tourism doesn’t from the traditional sightseeing tourism. The difference
between them is not only reflected in the purpose of consumption, consumption behavior,
consumption level, the consumption pattern, and also reflect in the difference of tourism
product, service, service system and quality demand. Therefore, to develop leisure
178 RISTI, N.º E5, 07/2016

tourism, we must first fully understand the objective situation of leisure tourism market
and the real needs of the market. Only in this way can we develop the suitable leisure
tourism products, guide and cultivate the consumer market. According to the forest
leisure tourism market survey research methods draw the following conclusion: visitors
choose tourism resources category to recuperate Resort and eco-tourism based, travel
time most of the election in a personal vacation, in favor of the majority of the people
above the two-day trip, half of the tourists willing to relatives and friends traveling; most
of the tourists every year in tourism spent more than 2000 Yuan, expenses mainly for
entertainment project participation and tourism commodity purchase, more than half
of the visitors to the tourism services in the convenient transportation and shopping
and entertainment aspects of concern; and little feature, lack of tourism attraction and
tourism projects in experience is discourage tourists travel the main reason. From it can
be seen that leisure travel more and more people like to participate in the experience of
the project market space is very large.
Tourism Intention classification Ratio (%)
weekend 18.4
Travel time holiday and vacations 23.1
Vacation 25.8
Personal travel 13.2
Travel agency organization 18.1
Way of traveling
Unit organization 16.5
Family or friends 50.9
<500 9.9
500~1000 24.1
Travel expenses
1000~2000 39.7
>2000 26.3
Attractions tickets 8.1
Entertainment project 43.8
Tourism service concerns
Dining room and board 19.1
entertainment 31.1
Table 4 – Tourism intention characteristics

In the investigation of the purpose of tourism, 45.3% of the tourists with the purpose of
leisure, the proportion was significantly higher than the tourism, and the other tourist
destination with a larger distance. At the same time, the result of this example is basically the
same as that of the tourist destination (Table 5). This shows that the demand for leisure and
vacation is constantly improving, and forest tourism is indeed an attractive leisure resort.
Tourist Leisure sightseeing Exercise visiting Conference shopping

destination vacation fitness family business
Percentage (%) 45.3 38.6 8.7 1.9 2.9 1.5
number 227 193 43 9 15 8
Table 5 – Tourist in the purpose
RISTI, N.º E5, 07/2016 179

In the setting of forest leisure tourism product type, 31.5% of the people want to
experience leisure sports, followed by leisure, theme park, leisure tourism and other
types of industries. People of this kind of different ways of the degree of the demand
is different, the difference is obvious, show a wide range of leisure tourists, visitors to
the leisure product selection is more likely to participate in the experience type. From
these two aspects to consider, you can set the product toward leisure, freedom of form,
a variety of ways to develop, to participate in the experience of products based, while
focusing on creating an atmosphere of leisure tourism. From the trend of should extend
the product in time, at the same time according to the needs of tourists difference set up
multiple topics and the contents, types, number of participants, in the way to optimize
the combination of, rich products, establish complete product package namely multi day
tour of leisure tourism products.
5. Conclusions
Forest tourism has been more and more people attention. The leisure tourism is being
accepted by more and more people. Leisure tourism has gradually become the main trend of
the development of China’s tourism industry. At the same time, it gathers the large number
of geology, sky, water, culture many different types of tourism resources. Forest tourism
product in a broad sense, that is a general concept, it is not only for a certain purpose
and provide material products and service products, but to in order to achieve a tourism
activities, tourists of tourism supply and in the direction set to eat, live, transportation,
travel, shopping, entertainment in the combination of various physical products and
service products from the perspective of tourism, forest tourism product is forest tourism
in order to obtain material and spiritual satisfaction by spending a certain period of
time, cost and energy of a forest tourism experience. The narrow forest tourism product
is in the area of forest tourism with good forest landscape and ecological environment
development, to meet the needs of the tourism product individual tourists or a variety of
forest tourism demand, it has natural wilderness, participation and the experience, interest
and knowledge leisure recreational, popular science and educational characteristics.
Application of GIS spatial analysis technology of forest tourism resources is a kind
of feasible method and technology, forest tourism resources are products of the
forest and its dependence on forest resources, forest has obvious spatial distribution
characteristics, and it controls the forest ecological benefits spatial distribution, studies
in the past due to backward technology, rarely involving spatial distribution of forest
ecological benefits. Today has been in the digitalization, visualization of the information
age, scientific and accurately quantify and visualization of ecological benefits provides
technical conditions. The research results show that, the GIS spatial analysis technology
and its application in the research of forest tourism resources has very broad prospects
for development.
References
Ashutosh, K., Shikhar, D. (2015). Identification of suitable sites for organic farming
using AHP & GIS. The Egyptian Journal of Remote Sensing and Space Science, 18,
181–193.
180 RISTI, N.º E5, 07/2016


(14), 1–15.
Kostas, K., Nikolaos, A. (2013). Virtual Fire: A web-based GIS platform for forest fire
control. Ecological Informatics, 16, 62–69.
Mattheus, C. R., Norton, M. (2013). Comparison of pond-sedimentation data with a
GIS-based USLE model of sediment yield for a small forested urban watershed.
Anthropocene, 2, 89–101.
Mukesh, S., Vít, V. (2015). Land use/cover disturbance due to tourism in Jeseníky
Mountain, Czech Republic: A remote sensing and GIS based approach. The Egyptian
Journal of Remote Sensing and Space Science, 18, 17–26.
Petr, V., Jan, K. (2015). GIS tool for optimization of forest harvest-scheduling. Computers
and Electronics in Agriculture, 113, 254–259.
Romain, L., Jagannath, A. (2015). R as a GIS: Illustrating Scale and Aggregation
Problems with Forest Fire Data. Procedia Environmental Sciences, 27, 66–69.
Sandro, S., Isabella, D. (2013). Bioenergy production and forest multifunctionality:
A trade-off analysis using multiscale GIS model in a case study in Italy. Applied
Energy, 104, 10–20.
Shihong, D., Fangli, Z. (2015). Semantic classification of urban buildings combining
VHR image and GIS data: An improved random forest approach. ISPRS Journal of
Photogrammetry and Remote Sensing, 105, 107–119.
RISTI, N.º E5, 07/2016 181

Research on the Fluctuation of RMB Exchange Rate

based on Data Mining Method
Ying Ye, Yirong Ying
2180378643@qq.com
College of Economics, Shanghai University, Shanghai, 200444, China

Pages: 182–194
Abstract: Foreign exchange risk is an important part of financial market risk, it

will not only directly affects the capital flows and international balance of payments,
but also has a certain impact on the currency circulation and inflation. In this paper,
we present an exchange rate anomaly detection method by combing phase space
reconstruction and sequence anomaly. The result of this method could effetely
reflect the fluctuation of the RMB exchange rate. Through the structural equation
model, we find that effect of price level, interest rate level, central bank monetary
policy intervention , economic growth and the international balance of payments on
RMB exchange rate are - 0.19, 0.20, 0.69, - 0.45, - 0.58. The result proves that the
central bank’s monetary policy has the greatest impact on the RMB exchange rate,
the price level and the interest rate has small effect on RMB exchange rate.
Keywords: Artificial neural network, RMB exchange rate, purchasing power
parity, anomaly detection
1. Introduction
Exchange rate is the main regulating and controlling means and economic lever of
macroeconomic. It is not only a direct impact balance of foreign trade, capital flows and
the balance of payments, but also have a certain impact on the currency and inflation,
to a country’s finance, investment and allocation of resources occurred in. With the
collapse of the seventies of the 20th century “Bretton Woods system”, in a free floating
exchange rate system to promote, price fluctuations in the exchange rate becomes
more and more complex and dynamic, research on the exchange rate issue is facing
new challenges (Edwin,2000; Obstfeld,2000). Exchange rate behavior anomaly makes
for more speculation in foreign exchange markets in various countries, the monetary
policy and foreign exchange management has received the serious interference, some
even complete failure, and the foreign exchange market to bring huge risk (Samuelson,
1968; González, 2015). Foreign exchange risk is an important part of financial market
risk, if the monetary authorities to the lack of enough understanding, lax management
and control may lead to currency crisis, and can even cause more serious financial crisis.
In this sense, it is very important to accurately grasp the exchange rate behavior. It has
a strong theoretical and practical significance to monitor the risk of foreign exchange by
studying the abnormal exchange rate changes.
182 RISTI, N.º E5, 07/2016

In view of the fact that the RMB exchange rate in the international financial market and
Chinese market economy the importance, this paper studies RMB exchange rate behavior
and its internal operation rules and on the basis of reveal the underlying factors of the
abnormal fluctuations in the impact of the RMB exchange rate, its theoretical value and
practical significance mainly reflected in the following aspects (Ramsey, 1997; Sarantis,
2000; Catherine,2006). For RMB exchange rate behavior of the research provides a new
idea and vision. The basic analysis and technical analysis of the traditional exchange rate
research is mainly concerned with the characteristics of most of the data in exchange rate
time series. However, the noise of a person may be another person’s signal; the abnormal
data may contain very important information. Therefore, it is extremely necessary to study
the abnormal data (Brock, 1996). The general method of exchange rate time series analysis
is based on statistical theory of time series analysis method, it is also the main content for
the modern econometric theory, mainly based on the mathematical statistics model as the
foundation, through hypothesis model, parameter estimation and model testing means and
technology to obtain a mathematic model to describe the rule of exchange rate time series.
From the macro and micro perspectives to grasp the operation of the exchange rate of the
RMB exchange rate, the exchange rate policy and related policies to develop and adjust.
The traditional exchange rate determination theory to analysis of the relationship between
exchange rate and its related macro and micro economic variables from the economic
equilibrium framework, and time series analysis only from the changes in exchange rates of
the application of mathematics and logic method of the future changes in the exchange rate
for prediction and analysis.
2. Data mining and anomaly detection

2.1. Parameter method
Most of the econometric model method is based on all the observed values to establish
a global model to describe the behavior of the system, without taking into account the
complexity and nonlinearity of exchange rate behavior. To this end, more and more
scholars use more non parametric methods and nonlinear methods to predict exchange
rates, which include artificial neural network (ANN) model, wavelet analysis (WBA)
model, genetic algorithm (GA) model, and so on.Artificial neural network is consists of
a large number of simple processing elements are widely connected components of the
complex network, is achieved in the brain tissue of modern biological research on the
basis of the results of the proposed and used to simulate human brain’s neural network
structure and behavior. In 1993, Refenes and other first try to use the neural network
method to predict exchange rate movements. In recent years, more and more research in
this area, and in the traditional neural network prediction model has been improved in
many aspects, has made a lot of progress. Artificial neural network as a massively parallel
processing nonlinear system, it has a lot for the regulation of parameters and strong
nonlinear simulation operation ability, so that it can precisely describe the nonlinear
dynamic process, and prediction on the trend of changes in the exchange rate provides a
brand-new way of thinking
Wavelet analysis method is a kind of window size is fixed but its shape can be changed,
the time and frequency windows may change the time frequency localization analysis
method, known as “mathematical microscope”. At present, some scholars use wavelet
RISTI, N.º E5, 07/2016 183

Research on the Fluctuation of RMB Exchange Rate based on Data Mining Method
Figure 1 – The neural network
analysis method to analyze the exchange rate data, using wavelet multi-scale analysis of
the characteristics of the dollar and yen exchange rate analysis, found that some of the
inherent laws. Genetic algorithm is a computational model for simulating the biological
evolution process of natural selection and natural selection. It is a global probability
search algorithm based on natural selection and genetic variation. Genetic algorithm
is mainly used in exchange rate behavior research, and often combined with neural
network and other methods to build a model, which can effectively improve the accuracy
of short-term exchange rate forecasting, and achieve better forecasting results.
2.2. Anomaly detection method

Outlier mining can be formalized description: given a containing N data points, or
schema objects collection, and expected abnormal data number k, collection and the
rest of the data compared to the significantly different, abnormal or inconsistent with
the top k objects. On statistical methods based on the idea of statistical outlier mining
method comes from, the commonly used method is first for a given data set assuming a
distribution or the probability model, then according to the model, the inconsistent test
to determine the abnormal data. The method needs to know the data set parameters, the
distribution parameters (such as the mean, standard deviation, etc.) and the number of
abnormal data. In order to solve the shortcomings of statistical methods in dealing with
large data sets and multidimensional problems, the concept and method of distance
based outlier data is introduced in the research field of data mining. Distance based
approach is to detect anomalies by the distance between the data points or objects.
3. Abnormal data detection of RMB exchange rate time series

3.1. RMB exchange rate nonlinear correlation structure
At present, it is generally believed that the exchange rate system of observation time series is
nonlinear, and the traditional linear analysis method cannot describe the change of exchange
rate. Nonlinear correlation is the time series of the observed data with nonlinear dependence,
nonlinear correlation structure of the existence of time series to show a variety of complex
nonlinear characteristics. According to the theory of nonlinear dynamics mechanism of any
184 RISTI, N.º E5, 07/2016

time series can be as a determined by a nonlinear mechanism of input and output system, the
complexity of the system and the possible nature of the uncertain factors and so on, the time
series generated is difficult to understand. However, the nonlinear correlation structure of
time series may be generated in a deterministic nonlinear system, and it is also possible that
it is a stochastic nonlinear system.
Chapter empirical data from a sample of RMB against the dollar exchange rate daily
central parity data, the sample interval selected 2011 January 1st to September 30,
2015. With the advance of the reform process and the market, changes in the RMB
exchange rate flexibility and uncertainty also increasing, which makes the issue of
RMB exchange rate in many fields especially have attracted worldwide attention in the
field of international trade, and even become talks on economic cooperation between
China and other countries or trade groups often relates to the key issues. Due to exchange
rate fluctuations on the world economy and the international financial risks of the
reaction has a good timeliness and sensitivity, the academic community generally adopt
exchange rate fluctuations to study exchange rate behavior. In order to guarantee the
data integrity and the information is sufficient, the first order difference of the natural
logarithm of the original data sequence of RMB exchange rate is {Pt}:
Rt = ln Pt − ln Pt −1 (1)
Among them, Rt means t days of exchange rate volatility,Pt means t daily exchange rate
middle price,P t- 1 means the middle exchange rate of t-1. In addition, the application of
nonlinear test needs to use a stationary sequence, so the analysis of the original time
series before the need for a smooth processing. In this paper, the most commonly used
ADF unit root test on the sequence analysis of stable. The results show that the price
series of natural logarithm, then first-order poor branch of volatility series can satisfy
the stability requirement.
Figure 2 – Sequence of RMB exchange rate fluctuations

From table 1 can be seen two kinds of exchange rate series obviously does not obey normal
distribution, JB statistics values were much larger than any reasonable significance
RISTI, N.º E5, 07/2016 185

level of critical value: χ2 (2, 0. 01) =9.21, accompanied probability p 0. The kurtosis of
exchange rate fluctuation is greater than 3, skewness is not 0, showing a more obvious
“fat tail” feature. The stronger the end of the heavy tailed, the stronger the continuity
of the state, the more important it is to predict the trend of exchange rate. It further
shows that the clustering of exchange rate volatility, that is, a large range of fluctuations
concentrated in certain periods, while the small amplitude fluctuations are concentrated
in another period of time.
standard Maximum minimum

sequence mean value
deviation value value
The price of collateral 7.49101 0.49408 8.11280 6.80090
wave -0.00018 0.00085 0.00301 -0.00405
sequence skewness kurtosis JB statistic P value
The price of collateral -0.22929 1.41922 108.593 0.00000
wave -0.60618 5.72317 355.789 0.00000
Table 1 – Descriptive statistics of RMB exchange rate
3.2. Principle of phase space reconstruction

It is generally believed that chaos is a seemingly random phenomenon in the deterministic
system. We know that for stochastic systems, the value at any given moment from a given
point is not known at any point in the future, that is, the system is unpredictable in the
short term. The chaos and stochastic difference is that disorder of chaos is not simple, but
there is no obvious period and symmetry. It has rich internal levels of ordered structure,
nonlinear system, a new form of existence. According to the mathematical definition of
chaos, we find out the characteristics of the chaotic behavior of the nonlinear system. The
most commonly used test of chaos is the singular attractor and Lypaunov exponent of
chaos. Attractor is the convergence performance of the system, whether a stable dynamic
system is chaotic, in most cases can use an attractor to describe the system equilibrium
state or time limit path, a nonlinear system will eventually evolve into two different
attractors: indifferent attractor and strange attractor. The ordinary attractor has three
modes, the fixed point, the limit cycle and the integer dimension, which correspond to the
equilibrium, periodic and almost periodic motion of the non-chaotic systems. Everything
that is not a common attractor is called a strange attractor, which corresponds to a non-
periodic, seemingly random, disordered and steady state motion pattern in a chaotic
system. The C-C algorithm is mainly based on the correlation integral:
1 t   N  m N 
S ( m, N , r , t ) = ∑ C s  m, , r ,t  − C s  1, , r ,t   (2)
t s =1   t   t 
The description of the correlation of nonlinear time series, and the statistics for time
delayτand embedding dimension M as:
1 t
S ( m, r , t ) = ∑ C s ( m.r.t ) − C m ( 1.r.t )  (3)
t s =1  s 
186 RISTI, N.º E5, 07/2016

Select the maximum and minimum two radius r, defined as the deviation of the
corresponding value:
{ } { }
∆S ( m,t ) = max s ( m, r ,t ) − min s ( m, r ,t ) (4)
Usually m, N, R has a certain range of choices, for all the S (R, m, t) for the average:
1
S (t ) =
m ⋅ j
∑ ∑ S ( m,r,t ) (5)
m j
All the S (m, t) averaging:
1
∆ S (t ) =
m ⋅ j
∑ ∆S ( m,r,t ) (6)
m
Defined statistics:
Scor ( t ) = ∆ S ( t ) + ∆ S ( t ) (7)
It is desirable to the minimum value as the optimal delay time windowτw values, and
calculate the optimal embedding dimension m:
m = τ w / τ + 1 (8)
Using matlab7.0 software to over C-C algorithm programming, the simultaneous

estimation of RMB exchange rate volatility series delay parameter tau D and optimal
delay time window tau W, the maximum time delay parameter t value is 100, to prevent
the time delay caused by small information loss. Run the program to get a curve on the
t as shown in Figure 3.
Figure 3 – RMB exchange rate time series C-C algorithm curve
RISTI, N.º E5, 07/2016 187

3.3. Characteristics of forest leisure tourism products

The algorithm of data processing in accordance with data input sequence, in order to
reduce the influence of the orderliness of data can be pre given parameter m, run the
algorithm m times and before each run on data sets do random sort, in M, the final
results of the abnormal concentration of choosing the smoothing factor value to do the
most for I the set of exceptions. In addition, the length of the sequence S in the algorithm
is n=|I| and n. Algorithm can be described as follows:
{
I={ i1}; DS(I1)=0;
for each ij∈(Ij−Ij−1)
{
Ij=Ij−1∪ {ij};
dj=Ds(Ij)－Ds(Ij-1);
}
if max(dj)≤0
exit;
else p=label{ij};
for each ik∈(I−Ip)
{
dk0=Ds(Ij-1∪{ik})－Ds(Ij-1)
dk1=Ds(Ij∪{ik})－Ds(Ij)
if dk0－dk1≥dj
Ix=Ix∪{ij};
}
return Ix;
}
The marked point represents a set of action points that are generated by the abnormal
points in the time series, which indicate some kind of change before the exception occurs.
Figure 4 – Corresponding to the abnormal points in one dimensional sequence
188 RISTI, N.º E5, 07/2016

Here, the appearance of abnormal structure seems to be no obvious rule, some of the
smaller fluctuations of the points may also be the composition of this structure change
points, however, they must hide some important information. To cause the abnormal
structural changes, single from time series itself is more difficult to find the inherent
reasons, so in carries on the analysis to the abnormal patterns in the sequence, not just
rely on performance in time series, and should proceed from the angle of macroscopic
systems, from decided to exchange rate system, the factors to consider, only in this way
can we deeply understand abnormal data generating mechanism.
Figure 5 – Structure change point
4.1. Structural equation model
Abnormal behavior of the RMB exchange rate is not spontaneous, but by the impact of
the operation of the economic system, because of the relationship between economic
variables often complex structure by general statistical analysis methods often fail
to reveal the inherent relationship between them. SEM overcomes the deficiency of
traditional statistical analysis method, multiple regression, path analysis and verification
factor analysis statistical technology integration, can excavate complex structure relation
between the large amounts of data effectively, the reference exchange rate theory and
scholars point of view, combined with the actual situation of the Chinese economy,
summarizes that the macroeconomic factors that affect the RMB exchange rate changes
mainly in five aspects.
•• Price level: from the perspective of purchasing power parity, long-term
since the RMB exchange rate has been underestimated, exchange rate to the
purchasing power parity regression appreciation is stronger than by the inflation
rate high depreciation pressure, coupled with the balance of payments often
account and capital account surplus, so that the exchange rate returns to the
purchasing power parity is very natural. This can partly explain the renminbi in
the continued appreciation of the past two years.
RISTI, N.º E5, 07/2016 189

•• The level of interest rates: with the deepening of the economic crisis, domestic
inflation pressure has slowed, economic decline, in this context, lower interest
rates, despite the decline in the international balance of payments surplus,
but the appreciation of the renminbi against the dollar is still no devaluation,
and always maintain a steady rising trend. However, with the advancement of
the Chinese market, it can be predicted that the impact of interest rates on the
exchange rate will continue to strengthen. In addition, in recent years, China’s
short-term capital inflows for the purpose of arbitrage increased, it is possible
to strengthen the impact of the RMB interest rate on exchange rate movements.
•• Balance of payments: the direct cause of exchange rate changes in a country.
International balance of payment mainly includes three parts, one is current
account, including goods trade account, pay interest and profits, and service;
the second is the capital account, record all capital flows; three is the official
reserves. Affected by the impact of the financial crisis in 2008, China suffered
a large number of trade barriers, trade friction intensified, coupled with export
enterprises financing difficulties and financing costs rise and the pressure of
RMB appreciation and other various aspects of factors, resulting in the end of
the year exports decreased.
•• Economic Growth: economic growth is the main measure of a country’s
economic growth rate, economic growth rate of a country’s exchange rate
changes produced by impact is in many aspects, economic growth rate is high,
which means to increase revenue, improve the level of domestic demand will
increase imports from the country, leading to the current account deficit, will
make its currency down. The impact of China’s economic growth on the RMB
exchange rate is mainly through its impact on the import and export volume,
although this effect is not direct, its effectiveness can not be ignored.
•• Monetary policy intervention: the central bank intervention on the
exchange rate of RMB exchange rate decision has a very far-reaching impact,
this intervention mainly through monetary policy. Over the past few years,
the asset structure of China’s monetary authorities has been a huge change,
an important manifestation of the central bank’s foreign net assets in its total
assets accounted for the proportion of rising. The rapid deterioration of the
international financial crisis, in order to ensure economic growth and stabilize
the market confidence, the central bank has taken the moderately loose
monetary policy, this a series of adjustment makes inflation rate has been
effectively suppressed, the RMB exchange rate at a reasonable and balanced
level remain basically stable.
4.2. Theoretical model and its parameters

This choice of RMB exchange rate and economic growth factors, price level factors,
interest rate factors and factors of international revenue and expenditure and the
intervention of the central bank monetary policy factors as latent variables studied,
SEM is established for the purpose of reveal the structural relationship between the
latent variables. With regard to the choice of measurable variables, the theoretical
circles in the study of macroeconomic phenomena, the various variables which have
some research results.
190 RISTI, N.º E5, 07/2016

Latent variable type Latent variable Observed variables

X1 (consumer price index)
ξ1 price level
X2 (inflation rate)
exogenous latent X3 (Chinese interest rate)

ξ2 Interest rate level
variable X4 (Sino US spread)
ξ
Monetary policy X5 (M2 variable momentum)
ξ3 intervention
X6 (foreign net assets accounted for)
η1 RMB rate Y1 (exchange rate volatility)
η2 economic growth Y2 (industrial added value)
Endogenous
latent variable Y3 (foreign exchange reserves)
η International balance of
η3 Y4(actual use of foreign investment)
payments
Y5 (difference between import and export)
Table 2 – Variable definition table

On the basis of this theory, the factor analysis conducted a number of computational
experiments, investigates the correlation between the latent variables, determined
to do the price level, the level of interest rates and the central bank monetary policy
intervention as exogenous latent variables, the RMB exchange rate, economic growth and
the international balance of payments as internal students latent variables, ultimately
selected for the following research hypothesis to establish initial theoretical model, and
path diagram representation such as in Figure 6.
Figure 6 – Model path and its parameters
4.3. Evaluation model

In this paper, the evaluation of the research model is carried out from three aspects:
the test of parameters, the degree of fitting of the whole model and the evaluation of the
explanatory power of the model. In the overall model fit test, χ2 tests was also called the
absolute fit index, is the most basic indicators of overall assessment model:
RISTI, N.º E5, 07/2016 191

(
χ 2 = ( N − 1 ) min F ~ χ 2 1 / 2 ( q ( q + 1 ) ) − k )
Through the study of the RMB exchange rate structural equation model, the structural
relationship between the RMB exchange rate and the economic latent variables is found,
and the role of the exchange rate is found. According to the estimated results of the final
model, the direct effects, indirect effects and the total effect of the latent variables on
the RMB exchange rate are summarized in Table 3. Indirect effect is the indirect effect
of exogenous variables on endogenous variables by affecting one or more intermediary
variables. When only a intermediary variable, the indirect effect of size is the product
of two path coefficient; when more than an intermediary variable, indirect effect is all
the exogenous variables of by one or several intermediary variables after the end in the
endogenous variables “chain of arrows” path numbers product.
route Direct effect Indirect effect Total effect

Price level→RMB exchange rate −0.24 0.05 −0.19
Level of interest rate→ RMB exchange rate 0.18 0.02 0.20
Central bank intervention→RMB exchange 0 −0.69 −0.69
rate
Economic growth→ RMB exchange rate 0 −0.45 −0.45
International balance of payments → RMB −0.58 0 −0.58
exchange rate
Table 3 – The effects of various latent variables on RMB exchange rate

Standardized effects of RMB exchange rate changes in the price level, the level of
interest rates, the central bank’s monetary policy intervention, economic growth and the
international balance of payments latent variables were: - 0.19, 0.20, and 0.69, and 0.45,
and 0.58, the coefficient of is that the impact is positive, and the coefficient is negative
indicates that its influence is negative, coefficient of absolute value is that the effect is
greater. According to the size of the force, this paper divides the factors into three categories:
1. The biggest impact on the RMB exchange rate is the central bank’s monetary
policy intervention, its direct effect on the RMB exchange rate is not obvious,
the impact of all the effects from indirect effects. As can be seen, the central bank
intervention mainly through the impact of economic growth and the international
balance of payments to achieve the purpose of affecting the RMB exchange rate
changes. Effect value is negative, and then the effect of intervention is mainly
from the stability of the RMB exchange rate.
2. The impact of the RMB exchange rate is more obvious on the international
balance of payments and economic growth factors. The international balance of
payments has a direct impact on the RMB exchange rate, which is determined
by the current account, capital account and foreign exchange reserves as a
comprehensive measure of the overall balance of the balance of payments.
Economic growth factors and the RMB exchange rate does not exist direct path,
but its indirect effect cannot be ignored, it is mainly through the impact of the
international balance of payments and thus affect the RMB exchange rate.
192 RISTI, N.º E5, 07/2016

3. Price level and the interest rates of RMB exchange rate effect is small, and the
effect mainly comes from the direct effect. This is also corroborated purchasing
power parity and the interest rate parity explanatory power in China is not
strong this conclusion.
5. Conclusions
Outlier data mining is an important branch of data mining, which is different from
other data mining methods. It is concerned with a small part of the data set. In many
applications, this kind of abnormal data often contains more important information and
knowledge than most of the data, so it has more research value. This paper discussed
in detail based on the actuality of the current exchange rate behavior research methods
and time series data mining, cut from the “abnormal” this new angle, using data mining
method to study the RMB exchange rate behavior. Through comparative analysis of a
variety of commonly used anomaly detection method, and considering the exchange
rate data of order and nonlinear, is proposed in this paper combining phase space
reconstruction with the sequential exception technique of outlier pattern detection
method of the RMB exchange rate series of anomaly detection, and from the macro view,
using SEM method to study the system of the various factors interaction mechanism,
and then find the deep reason of abnormal data of RMB exchange rate.
Traditional exchange rate determination theory to analysis of the relationship between
exchange rate and its related macro and micro economic variables from the economic
equilibrium framework, and time series analysis only from the changes in exchange
rates of the application of mathematics and logic method of the future changes in the
exchange rate are pre measured and analyzed. The former concerned about the reasons,
the latter is concerned about the results, there is no obvious intersection between the
two. In the analysis of exchange rate behavior, we should not only consider the micro
exchange rate price changes, but also should consider the impact of macroeconomic
factors, so the need to combine the two methods. Abnormal data mining using data
driven approach to explore the macro law, mining abnormal data hidden behind the
information, you can have a more comprehensive and profound understanding of the
mechanism of exchange rate.
Acknowledgments
This paper is supported by National Natural Science Foundation of China (71171128) and
Research Fund of Program Foundation of Education Ministry of China (10YJA790233).
References
Brock, W., Dechert, W. (1996). A test for independence based on the correlation
dimension. Econometric Reviews, 15(3), 197–235.
Catherine, K., Apostolos, S. (2006). Univariate tests for nonlinear structure. Journal of
Macroeconomics, 28(1), 154–168.
RISTI, N.º E5, 07/2016 193

Edwin, M., Raymond, T. (2000). Distance-based outliers: algorithms and application.

The VLDB Journal, 8, 237–253.
(14), 1–15.
Obstfeld, M., Rogoff,K. (2000). New directions for stochastic open economy models.
Journal of International Economics, 50(1), 117–153.
Ramsey, J., Zhang,Z. (1997). The analysis of foreign exchange data using waveform
dictionaries. Journal of Empirical Finance, 4(2), 341–372.
Samuelson, P. (1968). What classical and neoclassical monetary theory really was.
Canadian Journal of Economics, 1(1), 1–15.
Sarantis, N. (2000). On the short-term predictability of exchange rates: A BVAR time-
varying parameters approach. Journal of Banking and Finance, 30(8), 2257–2279.
194 RISTI, N.º E5, 07/2016

Study on Wireless Data Transmission System based

on GPRS and ARM
Mingyu Zhang *, Yuling Wang, ShuLan Gong
*mingyuofzhang@sdjzu.edu.cn
School of information & electrical engineering, Shandong Jianzhu University, Jinan 250101, China
Pages: 195–206
Abstract: With the development of wireless communication technology, wireless

data transmission technology is more and more used in environmental monitoring,
life and production. By analyzing ARM microprocessor, embedded real-time
operating system and GPRS, we proposed a wireless data transmission system based
on ARM and GPRS. This system combines the advantages of GPRS high network
coverage rate, and the low power consumption and strong practicability of ARM
microprocessor; it could be widely used in the field of remote data transmission.
The GPRS network and the Internet as the data transmission channel, the system
can realize the remote wireless transmission of data from the gateway node to the
monitoring center, and we give the gateway node and server hardware architecture
and the software realization process.
Keywords: Wireless data transmission, GPRS, ARM, system design
1. Introduction
Data transmission system is widely used in home security, equipment monitoring,
environmental monitoring, remote meter reading and other areas that require
unattended. Most of the traditional data transmission system based on Internet, optical
network, and other cable networks, common transmission medium are cable, twisted
pair, coaxial cable, the system technology is mature, stable performance, but cable
network in the building need dug trenches and wear, wiring, complex construction, high
cost of construction, network maintenance difficult(Ionel, 2012; Wancheng, 2012). In
the special geographical environment, such as ports, mountains, etc., the construction
of the cable network cannot be carried out. In addition, when the need to add new users
on the wired network, but also need to re wiring, system scalability is poor(ElFgee, 2014;
González, 2015). The rapid development of General Packet Radio Service (GPRS) solves
the problem of wired network cabling, which makes it possible to transmit remote wireless
data. GPRS is a GSM network to 3G broadband mobile communication, the transition
of 2.5G mobile communication system, it perfectly combines the data communication
technology and mobile communication technology, can make many users at the same
time, the use of channel resources, improve the utilization rate of wireless resources, in
the GSM to 3G transition process plays a very important role. GPRS data transmission
in a packet (Packet), the user’s burden is the cost of its transmission unit of computing,
RISTI, N.º E5, 07/2016 195

Study on Wireless Data Transmission System based on GPRS and ARM
not using its entire frequency band, and therefore relatively cheap (Liu, 2014; Fan, 2011;
Dehua, 2012). GPRS breaks through the GSM network can only provide circuit switched
communication mode, only need to the increasing of the corresponding functional
entities and the existing base station system is part of the transformation can be a packet
switched, this transformation of a small investment, data transmission rate. GPRS
network does not need an intermediary converter, so it is convenient to connect and
transfer. GPRS network has the following characteristics:
•• Make full use of the existing resources: the use of China Mobile’s
GSM wireless network covering the whole country, without wiring, can be
convenient, fast and low cost into a local user data terminal to provide remote
access network functions.
•• Transmission rate is high: GPRS transmission rate is up to 171.2Kbps
in theory, but the current rate can reach 20 ~ 30Kbps, due to the network
configuration and the limitation of the time slot of the GPRS module.
•• Access time is short: packet switching time is less than 1s, can provide fast
and timely connection, can quickly establish connection with Internet, and
improve the transmission efficiency.
•• Provide real-time online features: the user equipment will always be in
the connection and online state, which will make users access to the network has
become very simple and fast.
•• Billing according to the flow: GPRS users only send or receive data only
during the occupation of resources, users can always online, according to the
user to receive and send data packets to the number to the charges, no data
traffic transmission, the user even online but also do not charge a fee.
GPRS data transmission service can be applied to the entire low and medium rate data
transmission field. GPRS network can be used to maximize the use of existing public
network resources, reduce network construction costs, improve the quality of data
communication, and extend the communication distance. In this paper, the wireless
data transmission system based on ARM and GPRS technology, they are the current
rapid development of cutting-edge technology, theoretical and practical value is very
high (Zhang, 2006; Kang, 2006). On the one hand, the application of low power
consumption of ARM microprocessor and embedded operating system to realize
wireless data transmission between the data acquisition terminal and the control center,
save engineering wiring, reducing the damage to the environment, in line with China’s
construction of a resource-saving and environmental friendly society of national policy.
On the other hand, the GPRS network is easy to upgrade, high stability, wide coverage,
and communication costs low and not subject to geographical restrictions, etc., to reduce
the cost of network construction, increase the distance of communication, in line with
the strategic target of the construction of a conservation oriented society of our country.
2. Overall system architecture

2.1. System components
Based on arm and GPRS wireless data transmission system relies on the GPRS network
of China Mobile, the whole system is composed of monitoring terminal, including
196 RISTI, N.º E5, 07/2016

detectors and a data transmission module, GPRS wireless access and monitoring center
(PC, mobile phone). Wireless data transmission system, as shown in figure 1.
Figure 1 – Wireless data transmission system

Monitoring terminal is placed in remote monitoring, ARM microprocessor control sensors
to collect field data, and then use the GPRS module for data modulation, through the GPRS
network and wireless network operators network server to access the Internet. GPRS
wireless access, which is provided by Mobile Corporation GPRS network, it establishes
a seamless connection between the monitoring terminal and Internet, to achieve data
transmission between the monitoring terminal and the monitoring center. The monitoring
center is composed of one or more PC machines, the PC machine is connected with the
Internet, which is used to receive the data transmitted by the monitoring terminal and
process the data. Staff can monitor the site immediately to understand the situation.
2.2. GPRS structure analysis

GPRS uses the existing GSM network base station subsystem(BSS), but corresponding
updates to its software and hardware; BSS and service GPRS at the same time supporting
the Serving GPRS support node node(SGSN) is connected and need to increase new
GB interface, and GB interface will lead to GSM network circuit switching equipment
cannot be used. Therefore, it is necessary to increase the data packet control unit (PCU)
and mobile station (MS) to support GPRS service. GPRS network structure, as shown
in figure 2.2. GPRS network in the GSM network based on the increase of PCU, SGSN,
gateway GPRS support node (GGSN) and other network equipment, the functions of
these devices are as follows:
•• PCU: PCU separates the data service from the GSM voice service quantity, has
increased the grouping function, realizes to the wireless link control, allows
many users to access the same wireless resources. PCU through the Gb interface
with the SGSN connection.
•• SGSN: SGSN for the user and GGSN to provide packet routing functions, it will
be the base station controller (Controller BSC, BaseStation) sent the packet sent
RISTI, N.º E5, 07/2016 197

to other SGSN or via GGSN external Internet network transmission. The SGSN
through designed Gn interface connected to the GGSN.
•• GGSN: GGSN is the key equipment of GPRS network connecting other
networks. GGSN sent to the SGSN packet data sent to the Internet network;
from the Internet logo has a mobile terminal address of the IP package, also
received by the GGSN, and then forwarded to the SGSN, and then transmitted
to the mobile terminal. GGSN through Gi interface with Internet connection.
Figure 2 – GPRS network structure
3. The system hardware design

3.1. ARM microcontroller
Arm is a general-purpose 32-bit RISC (reduced instruction set computer computer, RISC
processor with high performance, integrated very typical RISC structure characteristics,
has simple structure, small kernel, low power consumption and high performance
characteristics. RISC is a kind of instruction set corresponding to Instruction Set
Computer CISC (Complex) with complex instruction set. With the development of
information technology, CISC has introduced a lot of complex instructions, which lead
to more and more complex structure. 1979 University of California, Berkeley, David
Patterson Professor research have found that about 20% of a typical operational process
instruction is used repeatedly and completed 80% of, and another 80% of the instructions
are rarely used, he based on results are presented RISC concept and developed the RISC
I type machine and RISC type II machine. The main features of RISC and CISC are
shown in table 1.
198 RISTI, N.º E5, 07/2016

Comparative content CISC RISC

Command system complex simple
Number of instructions Greater than 200 Less than 100
Instruction format Greater than 4 Less than 4
Addressing mode Greater than 4 Less than 4
Instruction length Not fixed As long as
Program length Short longer
Software development time Short longer
Instruction usage frequency Relatively large difference Little difference
Accessible memory instruction Without limitation LOAD/STORE directive only
Instruction execution time Great difference Most completed in one cycle
Table 1 – RISC and CISC characteristic contrast

From the table 1, compared with the CISC, RISC software development is difficult, but
due to the small number of instructions, the average instruction execution cycle shorten
and improve the working frequency of computer; extensive use of general-purpose
registers, can greatly increase the speed of program execution.
Figure 3 – ARM micro controller
3.2. Serial communication circuit

LPC2210 has two serial ports, UART0 and UART1, respectively. UART1 adds a modem
interface based on UART0. RS232 is a full duplex communication standard, which can
perform data reception and transmission at the same time. The standard is used in the
early computer and MODEM link, MODEM and then through the telephone line for
remote data transmission. The RS232 standard includes a main channel and an auxiliary
channel, in most cases, the main channel is used, that is, TXD, RXD, GND three signals.
Programming by configuring the PINSEL0 register to determine whether to use the
Modem UART1 interface. Due to the electrical characteristics of the RS232 standard,
the standard logic 1 corresponds to the voltage of -5 ~ -15V, and the standard logic 0
RISTI, N.º E5, 07/2016 199

corresponds to the voltage of 5 ~ 15V. Therefore, when using the Modem interface, the
TTL level should be converted to the RS232 level to connect and communicate with
Modem. Communication, the use of hardware flow control, the use of SP3243E chip
level conversion. UART1 port level conversion circuit, as shown in figure 4.
Figure 4 – UART1 Level conversion principle
3.3. GPRS module circuit design

GPRS module is responsible for the connection of GPRS network and data transmission.
Module M23 range of working voltage is 3.3V ~ 4.5V, but working current is not stable,
the average current in the standby state less than 6mA, 230-260mA state average current
during a phone call, the maximum current in the GPRS slightly less than 250mA. At the
same time calls and GPRS transmission of the current maximum 1.7A. Therefore, in the
choice of voltage conversion chip to consider the voltage and current, in particular, to
pay attention to the current can meet the need, otherwise it will lead to GPRS module
can not read the SIM card, can not send data, etc.. Power supply voltage conversion chip
select National Semiconductor Corporation Ns LM2576, the chip output current up to
3A, the output voltage of DC3.8V, can meet the needs of GPRS module. Because M23
can only read the 3V and SIM 5V card, so SIM card power supply circuit needs a separate
power supply, the voltage conversion chip selected SPX1117 and set its output to 3.3V.
Therefore, the power supply circuit of the module divided into two branches, the path
lm2576 conversion 3.8V output, the path SPX1117 conversion for 3.3V output. Input
voltage from the external +12V DC regulated power supply, but also to ensure that the
DC regulated power supply is greater than the average current 500mA, or plug in the
GPRS card after the SIM module will not read the SIM card. Between the core board and
M23 through the serial port for communication, generally using RS-232 port, LPC2210
through the UART1 to send AT commands to operate the module. Due to the need to
use the GPRS M23 function, data flow control to take the hardware, so we should use
all the serial signal pins, including DTR, RTS, DCD, CTS, RXD, TXD, RI and GND. M23
and LPC2210 communication need conversion, the conversion chip using ICL3238, as
shown in figure 5.
200 RISTI, N.º E5, 07/2016

Figure 5 – M23 serial level conversion
3.4. Flash interface circuit

Flash can be lost in the case of long-term storage of data, is a non-volatile memory,
storage characteristics equivalent to the hard disk. The device has the advantages of
small size, low power consumption, strong earthquake resistance and the like, and is
the preferred storage device in the embedded system. Flash memory is divided into
Flash Nand and Flash Nor. Flash Nor in the implementation of the program can be run
directly on the Flash Nor and do not need to copy the program to RAM, read fast, high
transmission efficiency, the interface is simple, but the price is higher. Flash Nand to
block as a unit of storage, cell density and high storage density, large capacity, write
and wipe out the speed of fast, erase unit is small, and the price is low. However, stored
procedures need to read the program to read the memory to run, manage complex,
and need a special interface. System Flash Nor selection of SST company production of
CMOS multi-function Flash chip SST39VF160, the storage capacity of 2M, data width of
16, the use of parallel interface. SST39VF160 pin function, as shown in table 2.
Symbol Pin name Pin function

A19—A0 Address input Block erase, A15 - A19 is used to select blocks.
DQ15—DQ0 Data input/output Sector erase, A11 - A19 used to select sectors;
Write cycle data within the internal latch OE# or CE# for high
CE# Chip enable output for the three state read cycle output data, write cycle
input data
OE# Output enable CE# start work for low time start
WE# Write enable Gating signal of data output buffer
VDD Power Supply Control write operation
V SS land Supply power supply: 2.7 - 3.6V
Table 2 – SST39VF160 pin description
RISTI, N.º E5, 07/2016 201

4. System software design

4.1. Hardware driver layer
Hardware drivers are usually written in assembly or C language, the program on the part of
the operation to function for the package, for the upper level program call. System I2C bus,
UART1, A/D interface and Flash Nand and other parts need to write driver. I2C bus and
uart1 drivers use with LPC2210 development of I2C bus and modem driver middleware,
application must pay attention to I2C and URAT1 vectors of interrupt initialization, and add
the interrupt service routine. LPC2210 has 10 8 bit A/D interface, using A/D control register
(Control Register A/D, ADCR) to select the working mode, A/D data (Data Register ADDR)
(A/D) to save the collected data. Before using the ADC module first through a pin setting
registers set the channel as a function of the A / D, and then set the channel mode, channel
conversion, conversion clock, conversion of the control mode using the adcr register etc. and
start the ADC conversion. By querying the highest level of the ADDR register to determine
if the AD conversion is complete, if the highest ADDR bit 0 is converted to an end. A/D
converter basic clock by the ultra large scale integrated circuit peripheral bus (Peripheral Bus
VLSI, VPB) clock to provide, and gradually adjust it to 4.5MHz, the clock frequency formula
as shown in formula:
FPCLK
=
CLKDIV −1 (1)
FADCLK
Bit15~bit6 ADDR for the data bits, using the data must be converted before conversion
formula as shown in formula (2):
VRe f
=
V AD × ADDR (2)
CA
In the operation of the Flash Nand usually need to load the memory technology driver
(Drivers MemoryTechnology, MTD), which is an important difference between Flash
Nand and Flash Nor. Flash MTD Nand is mainly to read, write and erase, the main
driver of the work is the preparation of ReadOnePage (), WriteOnePage (), EraseBlock
() three underlying functions. The operation of the K9F2808 is achieved by writing the
corresponding instruction data to the device instruction register. K9F2808 instruction
and timing definitions see table 3.
Busy time acceptable
function First cycle Second cycles
command
Read data register (data area) 00H/01H 
Read data register (free area) 50H 
Read device ID 90H 
reset FFH  O
Write data 80H 10H
Block Erase 60H D0H
Read state 70H  O
Table 3 – K9F2808 instruction and timing
202 RISTI, N.º E5, 07/2016

In the process of data transmission reliability, accuracy and data loss problems affected
by various aspects such as: GPRS network signal strength, geographical environment,
GPRS network to transfer protocol and other factors will cause the data there is a certain
deviation. At present, almost all of the mainstream operating system and more and more
GPRS the communication module (such as GTM900, SIM100 etc.) have built-in TCP /
IP protocol stack, and therefore do not need to write your own TCP / IP protocol reduces
the software part of the workload. Only the selected data transmission protocol used
in wireless transmission are analyzed in the GPRS network transmission control layer
consists of TCP and UDP two transmission protocol.
Figure 6 – TCP/UDP data transmission
4.2. μC/OS-II operating system transplant

μC/OS-II is a complete, concise, portable, can be cut, cured preemptive multi task real-
time embedded operating system can be used for a variety of microprocessor systems,
is designed for embedded applications in the design of a embedded real-time operating
Figure 7 – μC/OS-II operating system
RISTI, N.º E5, 07/2016 203

system. Most of the operating system code is written in ANSI C language, users as long
as the standard ANSI C cross compiler, assembler, linker, software tools, you can C/
OS-II transplanted to the development of products. C/OS-II has the characteristics
of high efficiency, small footprint, good real-time performance, high scalability, fixed
task execution time, stability and reliability, the kernel can be compiled to 2KB. C/OS-
II kernel is the basic functions of a based on scheduling priority preemptive real-time
kernel, task management, task scheduling, time management, memory management
and task communication and synchronization.
4.3. Design of user application layer program

Application layer program, located in the top layer of the system, users according to the
system requirements set a series of tasks to modular programming, CPU according to
the priority of these tasks turns service to these tasks in a. This mode of operation is very
much like the former background system, but there are many background tasks. Multi
task operation makes application modular, CPU utilization rate is high, the application
program is easy to design and upgrade. When the program is run first initialize the
system hardware, and then calls the function OSInit) (initialization operation system of
all variables and data structures, and then call the function OSTaskCreate) (create user
task, the call OSStart () start multi task. At this point the operating system takes out the
highest priority task from the task ready list.
File manipulation tasks of the main function is to wait for other tasks of reading to write the
message, once it receives the message is added storage devices, create a file, looking for a
location in memory and according to the information content to implement different read and
write API function. File operations using the ZLG/FS file system related functions, according
to the different file operation commands to read and write data. File manipulation task is:
first wait for the read and write command, the operating command is added at the bottom of
the flash drive, or continue to wait; then according to the command corresponding operation;
at last, delete the bottom of the flash drive. After executing the file write operation, call the
OSAllCachewriteBack function (), if the Cache full time still did not execute the command, the
user operation failed. The key code of the file operation task is as follows:
void TaskFile (void *pdata)
{
HANDLE FHandle;
pdata = pdata;
while(1)
{
comd=(uint8 *)OSMboxPend(FileBox,0,&err);
OSAddFileDriver(FlashCammand);
FHandle = OSFileOpen(FileName, “rw”);
OSFileSeek(FHandle,0, SEEK_END);
switch(comd[0])
{
case DISK_RDAD:
OSFileRead(ReadBuf,256,FHandle);
break;
204 RISTI, N.º E5, 07/2016

case DISK_WRITE_I2C:
OSFileWrite(WriteFileData,sizeof(I2C_Data),FHandle);
break;
case DISK_WRITE_Adc:
OSFileWrite(WriteFileData,sizeof(Adc_Data),FHandle);
break;
default:
break;
}
OSFileClose(FHandle);
OSAllCacheWriteBack();
OSRemoveFileDriver(0);
}
}
5. Conclusions
This paper analyzes the research status of wireless data transmission system, studies the
ARM microprocessor, embedded real-time operating system and GPRS technology, and
proposes a wireless data transmission system based on ARM and GPRS network. On the
hardware of the system, the LPC2210 is chosen as the main controller, the LM75 acquisition
temperature information is selected, the Flash Nor and Flash Nand are extended respectively
for storing program and data, and the GPRS module M23 is used as the data transmission
device. In the hardware circuit design, choose LPC2210 ARM7 microprocessor as the main
controller, the expansion of NOR flash, NAND flash were used to store programs and data,
selected GPRS module M23 embedded TCP/IP protocol stack as data transmission module,
through at commands to operate the TCP/IP protocol to connect to the Internet, reduced
the difficulty of programming, shorten the development cycle.
In the software design, the μC/OS-II real-time operating system and ZLG/Fs file system
transplant to the LPC2210, when first initialize the hardware and operating system, and
then start acquisition program, and the collected data is saved to NAND flash. System
in the transmission of data to detect whether the network is clear, if the network is clear
from the Flash Nand to remove the data and through the M23 module GPRS to send
data to the remote control computer. The system is stable, reliable and low cost, and
the test shows that the system can meet the actual needs of wireless data transmission.
References
Dehua, W., Pan, L. (2012). Water Quality Automatic Monitoring System Based on GPRS
Data Communications. Procedia Engineering, 28, 840–843.
ElFgee, E., Arara, A. (2014). Technical Requirements of New Framework for GPRS
Security Protocol Mobile Banking Application. Procedia Computer Science, 37,
451–456.
Fan, D. (2011). The Design and Implementation of a Wireless Remote Image Monitoring
System Based on the ARM and GPRS. Procedia Engineering, 24, 83–89.
RISTI, N.º E5, 07/2016 205


(14), 1–15.
Ionel, R., Vasiu, G. (2012). GPRS based data acquisition and analysis system with mobile
phone control. Measurement, 45, 1462–1470.
Kang, J., Nath, B. (2006). A study on switching voice traffic seamlessly between GSM
and GPRS cellular networks. Computer Communications, 29, 1079–1093.
Liu, Z. (2014). Hardware Design of Smart Home System based on zigBee Wireless Sensor
Network. AASRI Procedia, 8, 75–81.
Wancheng, X. (2012). Monitoring System of Power Line Icing Based on GPRS. Physics
Procedia, 33, 705–711.
Zhang, Y., Soong, B.(2006). A dynamic channel assignment scheme for voice/data
integration in GPRS networks. Computer Communications, 29, 1163–1173.
206 RISTI, N.º E5, 07/2016

Research on the Appraisal Index System of Real

Estate based on BP Neural Network
Jia Liu1,*, Xiaode Zuo1, Mingjian Xu2
*ampmbm@163.com
1
Management school, Jinan University, Guangzhou 510000, China
2
Guangzhou kanyi trade ltd, Guangzhou 510000, China
Pages: 207–216
Abstract: On the basis of the theory of real estate appraisal, this paper introduces
various methods of real estate appraisal. In reference to the previous experience and
the characteristics of examples, we choose 8 key factors to establish a price influencing
factor index of residential real estate. At the same time, we point out the defects of
the traditional market method, which is the degree of subjective assessment is high
in determining the weight of influencing factors. In order to overcome the defects
effectively, the BP neural network is established to train the index set. Through the
data forward propagation and error back propagation, the experimental results show
that the model has a good effect on the data fitting.
Keywords: Appraisal index, BP neural network, real estate, predicting price
1. Introduction
In recent years, the total investment and construction area of real estate development have
increased significantly, which is accompanied by the prosperity and development of the real
estate market, especially in the residential property market. In order to suppress caused
real estate prices too rapid growth to return to rational, the state promulgated the a series of
policy, strengthen the supervision on the market, curb excessive demand and standardize
order in the real estate market, but also for the real estate valuation industry provides a
healthy, transparent social atmosphere(Mark, 1991; West, 2000). With the growing
maturity of the real estate market, real estate sale, exchange, gift, leasing, etc. economic
activities are on the increase, these economic activities must take place to professionals on
the value of the real estate of assessment on the basis of the results, which is in line with the
provisions of the law of our country, but also can guarantee the fair, protect the interests of
the parties is not compromised (Ionel,2012;González,2015). This objective to promote the
development of real estate appraisal industry, also makes the real estate appraisal theory
has a strong practical value. The most commonly used methods of real estate appraisal
market law, income method, and cost method. Because of the maturity of the residential
real estate market, the market law is the best choice for the assessment of real estate.
However, real estate and ordinary goods are different, and its non-mobility makes every
case of real estate is unique, which gives the assessment increased the difficulty of the work.
Therefore, the use of market approach, the key is to large collection and assessment of
RISTI, N.º E5, 07/2016 207

Research on the Appraisal Index System of Real Estate based on BP Neural Network
the real estate is similar to the real estate transaction examples, using a modified version
of the influencing factors of real estate price, the value of property in the valuation date
to be assessed(Qi, 2005;Hu, 2006). However, the traditional market method has some
limitations in reflecting the nonlinear relationship between real estate price and real estate
price. Therefore, this paper establishes comprehensive real estate price influence factors
of evaluation index system, the artificial neural network is introduced to the traditional
market research method, using MATLAB software to establish the model, through the
establishment of a real estate evaluation model of artificial neural network (ANN) based
on, and for the real estate evaluation provides a new operation method.
Real estate assessment of the principle, the so-called real estate assessment, is to
consider the specific assessment of the purpose of professional personnel, the real estate
at a certain point in price estimates(Sun, 2010). When evaluating the assessment, the
assessment must be subject to the constraints of the industry standards. These principles
include the most effective use of the principle, the principle of legality, the principle of
price point, the substitution principle, the principle of fairness and prediction, and so on.
•• The most effective use principle: For the same real estate, it can be used for
industry, can also be residential, different purposes will have different benefits.
Make real estate to achieve the maximum benefit is the most effective use of. The
most effective use must be in compliance with the provisions of laws and regulations,
but also by the constraints of urban planning. In the real estate market developed
circumstances, the real estate can be used to achieve the most effective use of
competition(Meng, 2009). Therefore, the assessment of real estate should not only
consider the real purpose of real estate, if the real estate to achieve the most effective
use, should be the most effective use of the proceeds to assess the real estate prices.
•• Principle of legality: Real estate assessment of the primary premise is to check
whether the real estate property has a legitimate right to use and legal rights
and other legal rights. When calculating the net income of the real estate, the
operating purposes should be legal, for example, is not allowed to use the casino
in the mainland of china. When urban planning for commercial land, only by
commercial land revenue to assess, even if the residential land may be its best use.
•• Alternative principles: according to the principle of economics, the same
product with the same value, real estate is a commodity, but also follows this rule.
Namely in the unified market, the function of similar real estate prices should also
be similar. In determining the assessed value of the real estate, no matter adopt
what kind of method(Zhang, 2009). Finally, the need to and or similar market
transactions example, or real estate income examples, or is real estate development
examples were compared to be estimated in real estate prices do not deviate
significantly from these examples of real estate prices.
•• Principle of the price point: Price point is also known as the assessment
date, the real estate market price is fluctuating over time, lost the concept of
time, the results of the real estate assessment is meaningless. Therefore, in
the assessment of a real estate price, all the assessment data collected, the
information should be sorted to determine the assessment point in advance.
•• Principle of equity: Real estate assessment should be carried out in a fair,
impartial and objective principle, so as to ensure the legitimate rights and interests
of the transaction, while maintaining the authority of the asset appraisal work.
208 RISTI, N.º E5, 07/2016

•• Prediction principle: In real estate appraisal, not only to consider the real
estate in the past, and now the price, but also to form rational expectations of
future prices for real estate, in order to ensure the accuracy of the prediction,
prediction based on to realistic. The analysis of many problems in real estate
appraisal cannot be separated from the principle of prediction.
2. Artificial neural network theory

2.1. Artificial neural network
The essence of the human brain is the most complex in the known universe, most
effective and most functional information processing system, for a long time, people
use biology, neurology, cognitive science, mathematics, computer science and a series
of scientific of neural network research, trying to explain the working mechanism of
human brain, understand the way the nervous system for information processing,
which is designed with similar to the human brain function of intelligent information
processing system. The artificial neural network is a neural network which can realize
the function of information processing based on the understanding of the structure and
function of the human brain. Artificial neural network is a complex network which is
composed of a large number of simple neurons, which is highly nonlinear and can be
implemented in complex logic operation and nonlinear relation. It has absorbed many
of the advantages of biological neural network, artificial neural network each neuron
accepts input of a number of other neurons, through treatment and influence the output
of other neurons, the mutual influence and restriction of mode to realize the nonlinear
mapping from input to output. Network performance is a simple superposition of the
overall performance rather than local performance. BP neural network has an input
layer, the output layer and the hidden layer of the network structure. With I input nodes,
a hidden layer and an output layer of the 3 layer BP neural network as shown below:
Figure 1 – Artificial neural network
RISTI, N.º E5, 07/2016 209

Artificial neural network (ANN) through their own network structure to realize the
memory ability of information, these memories are stored in the weights between
neurons, using a distributed storage, only from the single weight is cannot see the
information content, which makes the network has good fault tolerance. Neuron is the
most basic unit of artificial neural network, and it is a nonlinear element with multiple
input and single output. Neuron output will be affected by several aspects, including
input, network weights and thresholds.
Figure 2 – single neuron model
2.2. Hierarchical structure

The artificial neural network consists of an input layer, an output layer and a number of
hidden layers, the number of hidden layers can be equal to 1, or more than 1, according
to the actual need to determine. The hidden layer is 1 network, which is called the single
hidden layer neural network, and the rest is called the multi hidden layer neural network.
The input layer node and the hidden layer node and the hidden layer node and the output
layer node are connected through the right, the hidden layer and the output layer each
node has its own threshold. Actually, the input layer representation is the input signal
and the output layer and the hidden layer is parallel neuronal ensembles, hidden layer
and output layer of each node represents a neuron, when the input signal, followed by
through each hidden layer reaches the output layer. The activation function is the core of
an artificial neural network, and the ability of the network to solve the problem is largely
determined by the activation function used in the network. The threshold type activation
function is characterized by converting arbitrary input hard into 0 or 1 of the output:
 1, w∗P +b ≥ 0
A = f (w ∗ P + b) =  (1)
0, w∗P +b < 0
The output of the linear activation function is equal to the weighted input and the
threshold value is added, function as:
210 RISTI, N.º E5, 07/2016

A = f ( w ∗ P + b ) = w ∗ P + b (2)
Figure 3 – Data prediction and analysis
3. Real estate evaluation model based on BP neural network

In the BP neural network, the determination of the number of hidden layer nodes is
closely related to the performance of the network. If the hidden layer node number
is too small, will lead to the lack of available mapping information, at this time, the
network does not have the learning ability and the ability of information processing,
fault tolerance will also reduce the, is difficult to complete the training; if the implicit
layer node number, makes neural network structure complex, increasing the number
of iterations and longer training time, network is very easy to fall into local minimum,
but also reduce the capacity of network generalization. Also a most widely used in the
actual method is repeated training method, for the same sample, the same learning
rate, the same accuracy was achieved when the training step of the least number of
network hidden layer nodes is the optimal number of nodes. Here, in order to select
the optimal number of hidden layer nodes, this paper takes the last approach, based
on past experience; the number of nodes in the hidden layer is 10, 20, 30, 40, 50,
respectively. The error precision of network training is set to 1e-3, and the learning rate
is set to 0.01, and the standard BP algorithm is adapted to study. The following 5 graph
is the choice of the hidden layer of the number of nodes in the MATLAB approximation
of the interface.
Above figures and tables can be clearly seen MATLAB in training the network hidden
layer nodes for network training times and mean square error effect, compared with the
number of training and mean square error, can see, when the hidden layer node number
is 40, number of training least and are square error smallest, so that we set up in the
network implied layer node number 40.
RISTI, N.º E5, 07/2016 211

Figure 4 – The hidden layer node number 10 to 50
hidden layer nodes Training times Mean square error MSE (e-004)
10 6698 9.99733
20 6909 9.99993
30 3850 9.99997
40 1762 9.99555
50 2070 9.99881
Table 1 – Different number of hidden layer nodes network
212 RISTI, N.º E5, 07/2016

4. Empirical research
4.1. Data standardization
In the process of writing this article, through some area in a city real estate market,
intermediary companies and a number of units of visits to research, collected 10 cases
have been traded real estate case, due to the different indexes with different quantization
standard, the gap between the individual indicators greatly, such as the original data
without finishing the inputs to the neural network, cannot reflect the relations of equality
index, therefore, must be in front of the network input the data of standard finishing, the
normalization method for data standardization:
x − min
x′ =
max − min
index Prosperity traffic environment infrastructure renovation Apartment floor orientation Unit Price
layout
1 0.50 0.55 0.55 0.72 0.30 0.50 0.10 0.35 20673.00
2 0.70 0.77 0.75 0.80 0.85 0.80 0.30 0.65 23589.00
3 0.55 0.60 0.70 0.70 0.30 0.70 0.60 0.55 21621.00
4 0.70 0.77 0.76 0.75 0.80 0.80 0.60 0.90 23571.00
5 0.76 0.88 0.77 0.85 0.85 0.50 0.20 0.40 26250.00
6 0.65 0.65 0.75 0.73 0.90 0.50 0.20 0.90 23000.00
7 0.86 0.85 0.77 0.85 0.90 0.50 0.65 0.90 25329.00
8 0.70 0.77 0.74 0.80 0.90 0.80 0.30 0.30 23588.00
9 0.75 0.80 0.69 0.85 0.82 0.70 0.80 0.40 23928.00
10 0.85 0.91 0.80 0.85 0.75 0.90 0.90 0.90 28437.00
Table 2 – Factor score sheet
4.2. Operation model

Using newffh () function or newcf () function to establish the network, this function
can be network initialization, automatic selection of weights and thresholds. According
to the introduced in front of the BP neural network parameter selection method and
validation of the hidden layer activation function selected logsig and output layer
activation function purelin function selection, algorithm selection secant quasi Newton
algorithm trainoss.
net=newff(minmax(p),[3,1],{‘logsig’,’purelin’},’ trainoss’);
net.trainparam.show=100;
net.trainparam.epochs=10000;
net.trainparam.goal=0.001;
net.trainparam.lr=0.05;
[net,tr]=train(net,p,t);
RISTI, N.º E5, 07/2016 213

Figure 5 – The network training error curve

Test network:
Out=sim(Net,p_test);
Q=mse(T_test-out);
[M,B,R]=postreg(out,T_test);
Plot(p_test,T_test,’+’,p_test,out,’bo’);
Figure 6 – Test data fitting
Test and training network performance Y=sim(Net,p) ;

Plot(p,T,’+’,p,Y,’bo’);
214 RISTI, N.º E5, 07/2016

Figure 7 – Training data fitting
4.3. Result analysis

The data obtained after the test results of anti-normalization. From the above experiments
we can see that after training the BP neural network is simulated the relationship
between the various types of real estate price influence factors and real estate prices,
forecast and actual prices close to that BP neural network in function approximation has
excellent performance, after 72 steps of training reached accuracy requirements.
Serial Actual transaction Network forecast Actual error relative error

price price
28 24839.00 24756 -83 -0.00334
29 23774.00 23573 -201 -0.00845
30 23075.00 22867 -108 -0.00468
Table 3 – Comparison between actual output and desired output
5. Conclusions
In this paper, the real estate market development trend in recent years as the background
to the theory of real estate appraisal, detailed introduction of a variety of ways to assess
the real estate. The article points out the defects of the traditional market method, that is,
the weight of the real estate price influencing factors to determine the subjective degree is
relatively high, so as to affect the accuracy of the evaluation results. In order to overcome
this defect, this paper establishes BP neural network the index set for training, through
data spread forward and backward error transfer automatically correct the weights
and thresholds of the network, until the network accuracy to meet the requirements
so far. Selection of neural network structure and parameters were obtained from the
experiments, determine the number of nodes in the hidden layer, hidden layer of
RISTI, N.º E5, 07/2016 215

10,20,30,40,50 neural network is tested, and the results show that when the number
of hidden layer nodes is 40, the convergence speed and accuracy of the network are the
best; in determining the learning function, the traditional BP algorithm, were compared
comprehensively, when using the secant algorithm Newton optimal performance of the
network, so the number of output nodes of BP neural network used in this paper is 9, the
number of hidden layer nodes is 40, the output layer node number is 1 and the hidden
layer activation function was S function, the output layer activation function is linear
function, learning algorithm for Newton secant algorithm. Experiments show that the
model has a good effect on the data fitting.
The current residential real estate evaluation index system contains the content is
not clearly defined, can according to the characteristics of the case for an increase
in the number of evaluation index, in detail described in detail in real estate related
characteristics, improve the evaluation index system; expert scoring method with
a certain degree of subjectivity, its precision is not high, the accuracy of the model is
improved by increasing the number of samples.
References
(14), 1–15.
Hu, Z. (2006). Prediction of real estate price index based on neural network. Sun Yat-
sen University Graduate journal, 27 (2), 100–113.
Ionel, R., Vasiu, G. (2012). GPRS based data acquisition and analysis system with mobile
phone control, Measurement, 45, 1462–1470.
Mark, B., Fishman ,D. (1991). Using neural nets in market analysis .Technical Analysis
of Stocks& Commodities, 9, 18–21.
Meng, Q. (2009). BP neural network real estate project risk assessment research.
Business economy, 7,115–116.
Qi, X. (2005). Research on real estate price evaluation based on fuzzy neural network.
Journal of Wuhan University of Technology, 27 (11), 113–114.
Sun, A., Peng, C. (2010). Prediction of real estate price index based on Grey Theory and
BP neural network. Enterprise economy, 4, 124–126.
West, D. (2000). Neural network credit scoring models. Computers &Operations
Research, 27, 1131–1152.
Zhang, X., Zhou, F. (2009). Research on real estate appraisal model based on neural
network and its MATLAB implementation. Technical economics and management
research, 1, 23–25.
216 RISTI, N.º E5, 07/2016

Empirical Analysis on the Efficiency of Financial

Governance based on Computer-aided Data Mining
Chao Ma
* xingshengkey@sina.com
Dongling School of Economics and Management, University of Science & Technology Beijing, Beijing
100083, China
Pages: 217–226
Abstract: Enterprise has accumulated a lot of financial data in the process of

information development, but the commercial value of these financial data has not
been fully excavated and utilized. The main feature of data mining is to extract,
transform and analyze all kinds of financial data, making problem solving approach
more efficient. In this paper, the author constructs the appraisal model of financing
efficiency, and makes the empirical analysis by using DEA method. The input
index as total assets, asset liability ratio and total operating cost, results showed
that 45% corporate financing is effective. Therefore, the application of data mining
technology in financial analysis has a good analysis and evaluation results.
Keywords: Financial governance, data mining, asset liability ratio, technical
efficiency
1. Introduction
Enterprise has accumulated a lot of financial data in the process of information
development, but the commercial value of these financial data has not been fully
excavated and utilized. Traditional financial diagnosis just simply use financial and
statistical method of small amount of financial data analysis based on the evaluation,
but for enterprise for many years of historical financial data and industry tens of millions
of information processing is difficult, the current situation is the massive financial data
information about the potential depth is not enough in-depth traditional diagnostic
methods and using (Kaplan, 1996; Bebchuk, 2010; Nini, 2012). Data mining as an
efficient tool, its advantage is the ability to quickly from a large number of missing,
noisy, fuzzy and random data in the capture of valuable but not known information
(Shleifer,1997; González, 2015). On this basis, the unique business rules behind these
data, and these rules are summarized into a model. Modern enterprise financing theory
originated from the famous MM theory. This theory started the modern enterprise
financing efficiency theory of the first of its kind. A profound discussion is put forward
in the article, which has attracted the attention of many scholars in the theoretical
circle. The article points out that the enterprise market value and the choice of the
way of financing the enterprise is not related, that is to say the different financing
methods will not cause the difference in the market value of the enterprise (Aghion,
RISTI, N.º E5, 07/2016 217

Empirical Analysis on the Efficiency of Financial Governance based on Computer-aided Data Mining
2011; Acharya, 2013). This thesis puts forward a perfect market system, which is
put forward under the condition of strict hypothesis, that is, there exists a perfect
competition market and the existence of information symmetry. It can be known that
it is not possible to fully meet these assumptions in the real environment. So the theory
has great limitation in practical application (Edmans, 2011;Jenkinson, 2012). But if we
only consider its theoretical significance, the basic framework of modern enterprise
financing theory is put forward. It can be said that after the scholars for the theoretical
study provides a way of thinking. So many scholars have made further research on the
enterprise financing problems on the basis of the development of the theory on the
basis of the theory, which makes the theory has been extended.
Corporate finance refers to the behavior of the listing Corporation to raise funds from the
inside and outside of the company to realize the organization goal. As a kind of special
form of enterprise organization, listing Corporation’s financing behavior has its own
characteristics, but also has the characteristics of general enterprises. Comprehensive
analysis, listing Corporation financing behavior has the following characteristics.
With diversified financing target, listed company’s diverse financing goal is mainly
manifested in two aspects: one, if in order to enhance the value of the company for
the purpose of, so the financing behavior of listed companies sub goal can be diverse,
including obtaining long-term stable money supply, lower capital cost, adjust the
capital structure. With flexible financing, listed companies because it directly faces to
the capital market, and have a strong financing autonomy, while the listed company
can be flexible use of financial instruments, such as bank credit, bonds, and stocks to
raise capital. With complex financing decision-making mechanism, phase due to the
listed company’s financing decision as an important financial decision, usually need
after discussion of managers, and then to vote by the board of directors and the general
meeting of shareholders to through. The relatively large size of the listed company,
the financing way is flexible, the stakeholders involved in the financing behavior more
influence on society.
Figure 1 – Data mining system
218 RISTI, N.º E5, 07/2016

2. Data mining applications

2.1. Data mining
Data mining is to extract knowledge from a large amount of data. According to the
comprehensive analysis of the data mining system, the whole system structure of
the data mining is shown in Figure 1. Data management module in the chart is the
foundation, the function of the instruction and the model evaluation. Mining operation
module is the core of the algorithm, which involves: association analysis, decision
tree induction, Bias classification, regression analysis, a variety of clustering analysis.
The model evaluation module is the result of data mining. The main applications of the
model include association patterns, classification patterns, clustering.
2.2. Clustering method

The information obtained from financial analysis can provide decision basis for investors
and managers. The financial analysis mainly includes the analysis of financial data
distribution, the structure factor and the index ratio. The main feature of data mining
is that it can solve the problem of model and high efficiency by extracting, transforming
and analyzing all kinds of financial data. This requires enterprises to comprehensively
analyze the internal and external business development situation and trends, the
effective disclosure of financial information. Therefore the use of data mining technology
for the company’s financial analysis and diagnosis is efficient and important. This paper
on the basis of research results of scholars proposed data mining in the financial analysis
diagnosis mainly used in: one is the use of cluster analysis combined with financial
ratios analysis can be sample enterprises are classified according to the clustering and
grouping variables standard, can use comprehensive financial ratios, such as Z value
comprehensive scores. Second to find the strong association rules, namely in the
clustering analysis based on, enterprises will be different according to the condition of
financial divided into several classes, then, use association analysis algorithm, to find the
effect of each enterprise’s financial situation of a variety of factors, through the inductive
analysis, the strong association rules.
Figure 2 – cluster analysis
RISTI, N.º E5, 07/2016 219

Decision tree is a tree like structure that is similar to the flow chart, and it is an inductive
learning algorithm based on an example. Using the decision tree method to classify, can
be more intuitive to see the classification rules from the decision tree. The decision tree
method has the advantages of high speed, high precision and easy to generate results
when processing a large number of data sets. Decision tree technology can be classified
in this paper and the study of the case of the company’s financial situation. From the root
node to a leaf node of the recursion calculation and comparison tests in the process of
attribute values, by different attribute value down recursive judgment, and ultimately in
the leaf node of the decision tree obtained relevant conclusions by decision tree algorithm.
Decision tree algorithm can be divided into two categories: Based on information theory
(including the ID series algorithm and C5.0 algorithm) and the minimum GINI index
algorithm (including CARPT, SLIQ and SPWNT algorithm).
Figure 3 – Data processing and analysis
3. Financing efficiency evaluation model

3.1. Construction of input output index
In the selection of the input indexes of corporate financing efficiency, considering the
company gets financial capacity of input indicators should be the inclusion of asset size,
capital structure, financing cost. These three aspects respectively for the total assets of
the company, assets and liabilities rate, total operating costs.
•• Total company assets (Xl): The larger the size of the company, it is easier to
raise funds to meet the needs of the company’s production and operation. And
then the scale of the company needs to raise the cost of financing the lower the
company’s financing efficiency is higher. Therefore, the index represents the size
of the company’s assets, can reflect the company’s overall financing situation.
•• Asset liability ratio (X2): Asset liability ratio, also known as the financial leverage
coefficient, from the perspective of the capital structure reflects the impact on
the efficiency of corporate finance.
220 RISTI, N.º E5, 07/2016

•• Total operating cost (X3): Total operating cost is the company’s daily operations
and production activities, all costs must be put into operation. And the cost of the
company’s investment will affect the number of output, thus the size of the total
operating cost determines the size of the profit. The company’s financing costs
and the company’s ability to fund operations can be represented by the index.
In order to clearly evaluate the efficiency of corporate finance output indicators, this
paper from the company’s operating performance and efficiency of the use of funds to
reflect the two aspects. Companies to raise funds for the use of funds efficiency, net
assets yield as an indicator to represent, and operating performance is used to represent
the net profit and operating income of the two indicators.
•• Net profit (Y1): Net profit is the final result of a company’s production and
operation. The more often the net profit is, the better the performance of the
company’s production and operation performance is, and the less the net profit
is, the worse the performance of the company’s production and operation is.
The net profit is the main index to measure the performance of the production
and operation of the listing Corporation.
•• Total revenue (Y2): Operating income is the main business of the company, is
an important guarantee for the company to make profits. The indicators can be
integrated into the capital, the results of production and business activities.
•• Return on net assets (Y3): Refers to the profit margin and the average
shareholder’s equity ratio, the higher the index means the higher the income
of the investment, and the higher the index, the weaker the profitability of the
owners’ rights and interests. The indicator reflects the company’s ability to
obtain net income from its own capital. It’s an important indicator to measure
the profitability of listing Corporation.
3.2. Descriptive statistics

Through preliminary statistics and finishing this paper selected 20 manufacturing listed
companies in 2013 to 2015 annual financial report, to the selected input indicators (total assets,
assets and liabilities rate, business total cost) and output indicators (net profit, operating
revenues, net assets income rate), simple data processing. Results are shown in Table 1 and 2.
year
Input index statistical indicators
2013 2014 2015
Maximum value 31720299 31863318 28504460
Total assets minimum value 271030 226854 163688
average value 3590843 3365043 2984349
Maximum value 0.91 0.9017 0.9069
Asset liability ratio minimum value 0.3008 0.3203 0.3256
average value 0.59245 0.59774 0.62247
Maximum value 45708769 40624659 34344447
Total operating costs minimum value 225003 239829 196513
average value 3957860 3737180 3331653
Table 1 – Manufacturing listing Corporation DEA investment indicators
RISTI, N.º E5, 07/2016 221

year
Input index statistical indicators
2013 2014 2015
Maximum value 2075176 2022186 1372852
Net profit minimum value 2173 3104 3956
average value 181360 191583 172718
Maximum value 48097967 43480394 31337628
Total operating income minimum value 232039 9573121 200756
average value 4106747 3918217 3244227
Maximum value 0.966 0.4058 1.5996
Return on net assets minimum value 0.0036 0.0301 0.07
average value 0.14888 0.14602 0.27801
Table 2 – Manufacturing listing Corporation DEA output indicators
4.1. DEA evaluation model
First, assuming that there are n decision making units (DMU), any one of the decision
making units has an input vector X, an output vector Y. The decision making unit (DMU),
which satisfies the axiom of minimal and ineffective, can be produced:
 n n 
=T ( X ,Y ) ∑ λ j X j ≤ x , ∑ λ j Y j ≤ y  (1)
=  j 1 =j 1 
The following DEA model can be obtained:
   m − s + 
min θ − ε  ∑ si + ∑ si  
  =  i 1 =r 1  
 n
 s.t . ∑ xij λ j + si− = θx
 j =1
(2)
 n

 ∑i =1
yrj λ j − sr+ = yr

 θ , λ , si− , si+ ≥ 0
θ is a scalar, λ is the N×1 dimension constant vector. The θ is the DMU efficiency value,
0≤θ≤1. If these θ=1, then the DMU in the efficiency frontier, which indicates that the
DMU is in active state. Get technical efficiency (TE), through the comparative observation
and analysis of manufacturing listing Corporation, if the level of technical efficiency
222 RISTI, N.º E5, 07/2016

is relatively high, it shows that it will be a variety of integration into the ability of the
output in the middle level. By increasing investment, any enterprise can be proportional
to expand production scale, which is returns to scale invariant (CRS) assumptions
underlying meaning, namely technical efficiency value is not affected by the size of the
enterprise. This does not conform to the actual situation. Therefore, based on the basic
model of the previous DEA, we try to build the BCC model of financing efficiency.
 n n n 
T=( X ,Y ) ∑ λ j X j ≤ x , ∑ λ j Y j ≤ y, ∑ λ j =
1, λ j ≥ 0  (3)
= j 1 =j 1 =j 1 
And then we can get the following DEA model:
   m − s + 
min θ − ε  ∑ si + ∑ si  
  =  i 1 =r 1  
 n
 s.t . ∑ xij λ j + si− = θx
 j =1

 n
(4)
 ∑ yrj λ j − sr+ = yr
 i =1
 n
 ∑ λj = 1
 j =1
 θ , λ , si− , si+ ≥ 0

4.2. Calculation result

Under the above the DEA evaluation model on financing efficiency, through the use
of data envelopment analysis software MaxDEA5.2 on the collected data processing,
first of all to get 2013-2014 20 manufacturing listed company’s overall efficiency, the
technical efficiency (TE), pure technical efficiency (PTE) and scale efficiency (SE),
specific calculation results are shown in Table 3 shows:
TE Score PTE Score SE Score RTS
No
2015 2014 2013 2015 2014 2013 2015 2014 2013 2015 2014 2013
1 0.912 0.917 0.891 1.000 1.000 0.952 0.912 0.917 0.936 irs irs irs
2 0.842 0.904 1.000 0.851 0.912 1.000 0.989 0.991 1.000 drs drs -
3 1.000 0.838 0.915 1.000 0.839 0.916 1.000 0.998 0.999 - irs drs
4 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 - - -
5 0.849 0.873 0.867 0.859 0.877 0.875 0.989 0.996 0.990 irs drs drs
6 0.901 1.000 1.000 0.903 1.000 1.000 0.998 1.000 1.000 drs - -
7 0.749 1.000 1.000 0.947 0.990 1.000 1.000 0.791 1.000 irs - -
8 0.880 0.895 0.894 1.000 1.000 1.000 0.895 0.880 0.894 irs irs irs
RISTI, N.º E5, 07/2016 223

TE Score PTE Score SE Score RTS

No
2015 2014 2013 2015 2014 2013 2015 2014 2013 2015 2014 2013
9 1.000 1.000 0.985 1.000 1.000 0.985 1.000 1.000 1.000 irs - drs
10 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 - - -
11 0.842 0.904 1.000 0.851 0.912 1.000 0.989 0.991 1.000 drs drs -
12 1.000 0.838 0.915 1.000 0.839 0.916 1.000 0.998 0.999 - irs drs
13 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 - - -
14 0.849 0.873 0.867 0.859 0.877 0.875 0.989 0.996 0.990 irs drs drs
15 0.901 1.000 1.000 0.903 1.000 1.000 0.998 1.000 1.000 drs - -
16 0.749 1.000 1.000 0.947 0.990 1.000 1.000 0.791 1.000 irs - -
17 0.880 0.895 0.894 1.000 1.000 1.000 0.895 0.880 0.894 irs irs irs
18 1.000 1.000 0.985 1.000 1.000 0.985 1.000 1.000 1.000 irs - drs
19 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 - - -
20 1.000 0.933 0.894 1.000 1.000 1.000 1.000 0.933 0.894 - irs irs
Table 3 – Overall efficiency measurement results

According to the estimated results of the financing efficiency of the 20 manufacturing
listing Corporation before the calculation results, this paper is divided into the statistics
and classification, the results see table 4.
TE Score PTE Score SE Score

year Sample Firms
no Proportion no Proportion no Proportion
effective 7 35% 10 50% 8 40%
2015
Non effective 13 75% 10 50% 12 60%
effective 8 40% 11 55% 8 40%
2014
Non effective 12 60% 9 45% 12 60%
effective 9 45% 12 60% 11 55%
2013
Non effective 11 55% 8 40% 9 45%
Table 4 – Financing efficiency

The effective ratio of financing efficiency of China’s manufacturing listing Corporation
is relatively high. Among the 2013, 20 manufacturing listing Corporation, there are 9
companies are both pure technology effective and effective, accounting for 45% of the
number of all enterprises, while the financing efficiency of non-effective enterprises
accounted for 55%. In the 2014 year, 20 manufacturing listing Corporation, there are
8 companies are both pure technology effective and effective, accounting for 40% of
the number of all enterprises, while the financing efficiency of non-effective enterprises
accounted for 60%. It can be known, 2013-2014, the overall efficiency of the efficiency
of the overall financing of the listing Corporation to maintain a relatively high level, but
with a slight downward trend.
224 RISTI, N.º E5, 07/2016

5. Conclusions
The survival and development of enterprises mainly rely on their own management
ability, an enterprise wants to grow and develop, and it must have a high degree of
understanding of its own ills and potential crisis, corporate financial data, in particular,
need to pay attention to. With the development of enterprise, the importance of
financial diagnosis for enterprise management is becoming more and more important.
With the advent of the information age, information technology and application, the
enterprise has entered a period of rapid development, if the information is the lifeline
of the enterprise; the data is the blood flow. Enterprises in the development process
has accumulated a large number of business data, mining of these data will affect the
business managers in a timely and accurate way to get the information needed for
financial decisions. But at present, the commercial value of these business data has
not been fully excavated and utilized. Therefore, the use of data mining technology in
the financial diagnosis of a large number of financial data, from which to obtain an
effective early warning, planning information, is the development trend of risk oriented
enterprise financial management.
The perfect capital market and financial market system play a fundamental role
in the financing decision of the listing Corporation. Therefore, various measures
should be taken to optimize the capital market and financial market. First, continue
to promote the reform of the state-owned commercial banks. The key to improve the
safety and efficiency of bank loan funds is to promote the reform of the state-owned
commercial banks. Second, enhance the bank’s ability to control the camera. The
bank’s camera management mainly includes the prior reverse selection criterion,
the moral hazard control in the event and the ex post reorganization. Enterprise
more by the internal financing turn for the stock market financing, which means
that in the enjoyment of a wide range of social resources at the same time, and
also to accept from the external supervision of market, the financing of enterprises
not only to be more transparent in the data and also on the independence of the
challenge. Generally speaking, financing enterprises should expand financing
channels, such as bonds, multi joint role, to provide effective protection for the
development of enterprises.
References
Aghion, R. (2011). Incomplete contracts and the theory of the firm: What have we learned
over the past 25 years? The Journal of Economic Perspectives, 25, 181–197.
Acharya, V. (2013). Corporate governance and value creation: Evidence from private
equity. Review of Financial Studies, 26, 368–402.
Bebchuk, V. (2010). The state of corporate governance research. Review of Financial
Studies, 23, 939–961.
Edmans, A. (2011). Governance through trading and intervention: A theory of multiple
blockholders. Review of Financial Studies, 7, 2395–2428.
RISTI, N.º E5, 07/2016 225


(14), 1–15.
Jenkinson, T.,Mayer, C. (2012). The assessment: corporate governance and corporate
control. Oxford Review of Economic Policy, 8, 1–10.
Kaplan, R. (1996). Using the balanced scorecard as a strategic management system.
Harvard business review. 74, 75–85.
Nini, G., Smith, C. (2012). Creditor control rights, corporate governance, and firm value.
Review of Financial Studies, 25, 1713–1761.
Shleifer, A., Vishny, R. (1997). A survey of corporate governance. The journal of finance,
52, 737–783.
226 RISTI, N.º E5, 07/2016

Research on Fuzzy PID Control Strategy for

Brushless Direct Current Motor
Lili Jing 1, Hua Ge 1, Yang Nie 1, 2
* 15847418543@163.com
1
Department of Physics, Jining Normal University, Inner Mongolia, China
2
Digital Engineering Center, Communication University of China, Beijing, China
Pages: 227–237
Abstract: Towards the issues of traditional PID technology in brushless DC motor

control systems, such as bad control accuracy, poor immunity and so on. We discuss
an online self-tuning control strategy: Fuzzy PID control strategy. This control
strategy is based on the analysis and mathematical model of brushless DC motor
control principle, combining Fuzzy control and PID control technology with adaptive
factor, make it possible to adjust the controller parameters online automatically, thus
improving the performance of the controller to control brushless DC motor system.
This article also contains the fuzzy PID control strategy simulation experiment for
brushless DC motor system on the simulation platform by Model-Based Design
(MBD). Using this method, we can quickly build a fuzzy PID system simulation
model, and accomplish efficiently the system test, simulation and implementation.
Experimental results show that the accuracy and immunity of fuzzy PID control are
better than traditional PID, in brushless DC motor control systems.
Keywords: Brushless DC motor, Fuzzy PID control, Model-Based design
1. Introduction
The trapezoidal wave or square wave brushless DC motor (BLDC) is a kind of brushless
DC motor which is recognized universally (Guo, 2008; Abreu, 2015). The brushless
DC motor has a huge potential for development and application in defense, industrial,
office automation, automotive electronics, household and other areas, for its AC and DC
advantages. For the reason of wide application of the brushless DC motor and strong
market demand, the response to the national energy conservation and the call of the
green, as well as the protection to the motor supplies’ safety, stability and efficiency, the
research of high-precision control of brushless DC motor system, to improve product
performance, have important theoretical significance and market value. The fact is that
the traditional PID control technology for multi-variable, strong coupling and nonlinear
brushless DC motor control system can’t meet the demand. To solve these problems,
the paper uses fuzzy PID control strategy. It is the combination of the virtues of fuzzy
control technology and PID control technology, possessing both excellent suitability for
nonlinear and time-varying systems, but also to eliminate the steady-state error, and the
excellent characteristics of eliminating steady-state error.
RISTI, N.º E5, 07/2016 227

Research on Fuzzy PID Control Strategy for Brushless Direct Current Motor
2. Model of Brushless DC Motor and Speed Control Method

2.1. Mathematical Model of Brushless DC Motor
All In the premise of no influences of control performance, Experimental assumptions
are as follows (Hemati, 1990): stator of a brushless DC motor is completely symmetrical,
three-phase windings use Y-connected style, each phase winding difference at 120。,
and each winding’s resistance and inductance are the same. At the same time, magnetic
saturation and eddy current losses are ignored, winding are distributed evenly in the
smooth surface of the stator, rotor damping effect are Ignored, the wave of rotor magnetic
field is square wave, the wave of anti-voltage is trapezoidal wave. Without considering
armature reaction effect of stator windings and the mutual inductance between the
armature windings, symmetrical three-phase winding and the rotor reluctance do not
change with the position of the rotor. Its equivalent circuit and drive circuit (Junhyuk,
2004) as shown in Fig 1.
Figure 1 – Brushless DC Motor Equivalent Circuit and Full Bridge Drive Circuit
As we can see from the figure, A, B, C are the three-phase equivalent circuits of the
motor, V1-V6 are the six power switches, using two-two conduction mode. Use the
sampled position signal to determine which two phases conduct, then supply the power
through the inverter, forms the rotating magnetic field, so as to drive the rotor to rotate.
The three-phase brushless DC motor stator voltage equation expression as follows:
ua  R 0 0  ia   L − M 0 0  ia  ea 

      d    
= ub  0 R 0   ib  +  0 L−M 0  dt  ib  +  eb  (1)
 uc   0 0 R   ic   0 0 L−M   ic   ec 
In the formula(1), ua , ub , uc are terminal voltage of three-phase stator motor (V); ea , eb , ec

are counter electromotive force (V) in a three-phase motor winding; = R R= a R=b Rc
are three-phase motor winding resistance (Ω); ia , ib , ic are three phase current of stator
motor(A); =L L= a L= b Lc are the inductance of three-phase windings of the motor
228 RISTI, N.º E5, 07/2016

(H); M = L= ab L=
ac L=
bc L=ba L=
ca Lcb are mutual inductance between Three-phase
motor stator winding(H). Ignore the power switch operation transition process and
the inductance of armature winding. Electromagnetic power output of the motor is
as followed:
Pe = ea ia + eb ib + ec ic (2)

Ignore other losses of the motor rotor; the electromagnetic power can be completely
converted into kinetic energy of the rotor:
Pe = Teω (3)

The electromagnetic torque equation can be obtained by formula (2), (3).
Te = (ea ia + eb ib + ec ic ) ω (4)
In the formula, Pe is motor electromagnetic power; Te is motor electromagnetic torque

( N ⋅ m ); ω is motor machinery angular velocity ( rad s ). To construct a complete motor
mathematical model, the equations of motion of the motor are also needed:
dω
J = Te − TL − B ω (5)
dt
Where J is rotary inertia of the motor, TL is load torque ( N ⋅ m ),and B is damping

coefficient ( s −1 ).
2.2. Mathematical Model of Brushless DC Motor

According to the theory of DC motor, brushless DC motor speed can be expressed as:
U d I D − I D2 RΣ − 2∆UI D
n= (6)
2C e TD C T

In the formula, Ud is terminal voltage two-phase conducting windings. RΣ is total
resistance of the stator armature circuit. ID is the stator armature current. ΔU is voltage
drop of the power tube. Ce is internal voltage constant of brushless DC motor. CT is
internal torque constant of brushless DC motor. TD is motor torque.
Experiment with brushless DC motor uses the constant torque speed mode, internal
motor torque is constant, Motor power and speed are linear with each other. The speed
is changed by the change of internal motor armature voltage. Electromagnetic torque of
the motor does not change with the change in the motor speed.
Operating power of the motor varies with the change of the motor speed linearly at the
runtime. This speed control method has a high steady state performance, with a wide
RISTI, N.º E5, 07/2016 229

range of speed applications. This method can realize step less speed regulation, having a
small motor energy loss during the process of controlling speed. It can be widely applied
to the motor under the requirements of high-quality case.
3. Fuzzy PID Brushless DC Motor Control Strategy

3.1. Principle and Structure of Fuzzy PID Controller
The fuzzy PID controller of Brushless DC motor control system mainly focuses on its
characteristics of nonlinear and strong coupling. According to the system running
condition, use the fuzzy control thought to achieve the online nonlinear adjustment
of PID parameters timely. Always keep the good match of controller parameters and
controlled object, to obtain higher stability and quicker response of the system. These
virtues make it possible to meet the quality control requirements of motor supplies.
The basic principle of fuzzy PID controller is to input PID controller error e and error
change rate ec simultaneously to the fuzzy controller(Krishnan, 2009; Luo, 2009; Li,
2016), and to find out the fuzzy relationship between e, ec and PID parameters. The
aim is to achieve the real-time online adjustment of three PID parameters and amend
three parameters kp, ki and kd which are obtained by the fuzzy controller into the PID
controller, to meet the dynamic requirements of controller parameters in different e and
ec. Thus make it possible to achieve real-time control of the controller, which will enable
make the controlled object BLDCM to obtain a good dynamic and static performance
(Wang, 2003; Xia, 2009). Fuzzy PID controller structure is shown as the dashed box
in Fig 2. r (t) is a speed set value of brushless DC motor fuzzy PID control system, y (t)
is a output value of the BLDC fuzzy PID control system, n (t) is a the speed feedback
detection output value of the system, e (t) is the fuzzy PID controller input value, ec (t) is
the rate of change of speed deviation, u (t) is output value of fuzzy PID controller.
Figure 2 – Control Structure Diagram of the Fuzzy PID Controller
230 RISTI, N.º E5, 07/2016

3.2. Design of Fuzzy PID Controller

As can be seen from the fuzzy PID control structure, its control process can be divided into
fuzzy control and PID process control processes (Xu, 2011). Fuzzy PID control process and
control process can be seen as a serial process in each cycle. Fuzzy control process is forward.
PID control process is in the post. Their functions and objectives are different. Therefore,
the fuzzy PID controller can be designed according to the following two processes.
Fuzzy control process: firstly, the fuzzy controller input values speed deviation e and
error change rate ec converse and blur of quantized factor Ke, Kec, the corresponding
results are E and EC. Use three correction parameters △kp , △ki and △kd as output
values of fuzzy logic controller,△KP, △KI and △KD are corresponding fuzzy values
before clarity. To facilitate the design, take seven fuzzy factors to cover the fuzzy domain.
They are NB (Negative Big), NM (Negative Medium), NS (Negative Small), Z (Zero),
PS (Positive Small), PM (Positive Middle) and PB (Positive Big). The fuzzy domain of
E and EC is [-3, 3], and the fuzzy domain of △KP, △KI and △KD are [-1, 1]. The type of
membership function of fuzzy subsets was taken as Z-type, S-type and triangle-type.
Secondly, according to three different time control parameters and their interactions,
and practical experience and knowledge of operators and experts, we can get control
rules of fuzzy PID control parameters △KP, △KI, and △KD. When | E | and | EC| of
brushless DC vary, the rules are shown in Table 1.
Then according to the control rule, make △KP, △KI, △KD to be cleared, which are results
decided by Mamdani fuzzy inference engine. Use the maximum membership degree average
method(mom). Those above results are converted into a clear value △KP, △KI and △KD.
Then obtain PID corrected parameters △kp , ki and △kd by scaling factor Kup, Kui and Kud.
Finally, accumulate them with three initial parameters of PID kp0, ki0 and kd0. The
ultimate outputs of fuzzy logic controller are current values of parameter required for
conventional PID. The formulas are as follows:
kp=kp0+Kup△Kp=kp0+△kp (7)
ki=ki0+Kui△Ki=ki0+△ki (8)
|EC|
|E|
NB NM NS ZO PS PM PB
NB PB/NB/PS PB/NB/NS PM/NM/NB PM/NM/NB PS/NS/NB ZO/ZO/NM ZO/ZO/NS
NM PB/NB/PS PB/NB/NS PM/NM/NB PS/NS/NM PS/NS/NM ZO/ZO/NS NS/ZO/ZO
NS PM/NB/ZO PM/NM/NS PM/NS/NM PS/NS/NM ZO/ZO/NS NS/PS/NS NS/PS/ZO
ZO PM/NM/ZO PM/NM/NS PS/NS/NS ZO/ZO/NS NS/PS/NS NM/PM/NS NM/PB/ZO
PS PS/NM/ZO PS/NS/ZO ZO/ZO/ZO NS/PS/ZO NS/PS/ZO NM/PM/ZO NM/PB/ZO
PM PS/ZO/PB ZO/ZO/NS NS/PS/PS NM/PS/PS NM/PM/PS NM/PB/PS NB/PB/PB
PB ZO/ZO/PB ZO/ZO/PM NM/PS/PM NM/PM/PM NM/PM/PS NB/PB/PS NB/PB/PB
Table 1 –The Fuzzy Control Rules of ∆KP, ∆KI, ∆KD
RISTI, N.º E5, 07/2016 231

kd=kd0+Kud△Kd=kd0+△kd (9)
PID control process: the procedure uses the incremental PID control. After the sampling
period and the response factors are determined, simply enter the three measurement
bias during the process, the incremental amount of control. It represents the actuator’s
position change during two sampling time intervals. Compared with the position PID
control, incremental mode has much simpler computational algorithm, smaller amount
of calculation and it is easy to achieve by software programming, which is widely used
in the actual production. More specific implementation process of PID controller is
based on the speed error e and error change rate ec, and three fuzzy controller’s output
parameters ∆kp, ki and ∆kd generated by fuzzy inference. Finally obtain u is the control
amount of Fuzzy PID control system, through the proportional, integral, and cumulative
control processes. Note that during the PID control process, proportional, integral and
derivative control process may be not only a serial process, but also do parallel processing.
The premise is that it must be completed before the accumulation control process.
4. Modeling and Simulation of BLDC Fuzzy PID Control System

4.1. The Workflow of Model-Based Design
Model-Based Design (MBD) is innovating the way for engineers and scientists working.
In Model-Based Design, a system model is right on the center of the development process,
from requirements development, through design, implementation, and testing. Model-
based design gives us a complete product from idea to generate the code development
process. In MBD, a system model is right in the center around the development process,
from requirement’s development, through design, implementation, and testing. This
workflow is elaborated to create hardware and software partitioning, automatically
create hardware and software implementation code, and verify the hardware and
software implementations in the context of the complete system. It is not only
overcoming the defects of low efficiency and difficulty of meeting the requirements in
traditional methods of progress, but also to meet time-to-market and cost objectives.
MBD workflow is illustrated in Fig 3.
4.2. Overall System Simulation Model

In MATLAB / Simulink simulation software environment, according to the brushless DC
motor mathematical model and fuzzy PID controller structure, use a modular system
modeling idea to design and establish a general simulation model of fuzzy PID brushless
DC motor speed control system. As shown in Fig 4.
Fuzzy PID brushless DC motor speed control system simulation model includes two
parts. One is the brushless DC motor fuzzy closed-loop control system; the other is
performance evaluation system model. The brushless DC motor fuzzy closed-loop control
system includes three parts, including Speed given module n*, speed control subsystem
module F-PID and brushless DC motor model. The brushless DC motor model mainly
consists of DC power supply model DC, inverters the Universal Bridge, permanent
magnet synchronous motors and PWM pulse width modulator PWM Generator, pulse
232 RISTI, N.º E5, 07/2016

Figure 3 – The Workflow of MBD
Figure 4 – Fuzzy PID brushless DC motor speed control system
control unit controller. Performance evaluation system is established by IST2E, which

represent the controller’s performance of the control deviation according to “Optimal
Control”. Its empirical formula is:
∞
J 2 = ∫ t 4 e 2 ( t ) dt (10)
o
RISTI, N.º E5, 07/2016 233

4.3. Fuzzy PID Controller Simulation Model

The fuzzy PID controller is the core of the whole control system. According to the design
idea of the fuzzy PID controller of the brushless DC motor, set up fuzzy PID controller
model F-PID, in the work environment of Simulink, as shown in Fig 5. The whole
controller is mainly composed of two parts, the fuzzy controller module Fuzzy and other
arithmetic units of PID.
Figure 5 – Simulation Model of F-PID Module in the Fuzzy PID Controller

At the same time, fuzzy controller module Fuzzy is the core of the whole fuzzy PID
controller. Its main function is to provide three required parameters kp, ki, kd for the
fuzzy PID controller. The simulation model is shown in Fig 6.
Figure 6 – Simulation Model Block Diagram of Fuzzy Controller Fuzzy
4.4. Parameter Settings

Brushless DC motor control system simulation model parameter settings are shown in
Table 2.
234 RISTI, N.º E5, 07/2016

Stator resistance
The magnetic field flux 0.1848Wb
Rs=4.76 Ω
Stator inductance
Field width 120o
Ls=0.0085H
Brushless DC Motor
Number of pole pairs
Moment of inertia p=2
J= 0.0001051kg▪m2 rated power
P=5.5KW
rated power DC Ud=300V
Torque Reference TL=3N▪m
Amplifier Simulation Algorithm
G1=0.1
Gainl ode45
Table 2 – Parameter List of the BLDCM Control System

Use the traditional empirical method to obtain initial PID three parameters kp0=40,
ki0=1，kd0=0.0101, according to the principles of fuzzy PID controller aforementioned. To
realize the fuzzy control, calculate the input variables e and ec quantization factor Ke = 0.002,
Kec = 0.000007 according to the basic domain to fuzzy domain conversion formula. The scale
factors of output variables △Kp, △Ki, △Kd are that Kua = 0.065, Kui = 1.15, Kud = -0.00015.
4.5. Simulation Results

To verify the feasibility and effectiveness of the application of brushless DC motor speed
control of fuzzy PID control strategy in brushless DC motor speed control, apply directly
to the brushless DC motor speed control system. Make the necessary settings and model
parameters and embed the corresponding program file, make sure the system is running,
and record the results.
With load TL3N-m, white Noise and 0.05s sudden load torque disturbance, the waveform
of brushless DC motor system traditional PID control and fuzzy PID control are shown in
Fig 7. In traditional PID control, brushless DC motor speed overshoot 12.3%, the adjusting
time is 5ms. In Fuzzy PID control of brushless DC motor, speed doesn’t overshoot, settling
time is 4ms. After 0.1s, change the given speed, it changes to 1000r / min rapidly. Fuzzy
PID’s ability is significantly better than PID regulator’s. After 0.05s, give the sudden
disturbance torque, the former is slightly better than the latter, but not very obvious.
Figure 7 – Speed Waveform of BLDCM of Fuzzy PID Control and PID Control
RISTI, N.º E5, 07/2016 235

Compare the performance of Fuzzy PID controller and conventional PID controller,
simulation results are shown in Fig 8.
Figure 8 – Performance Comparison between Fuzzy PID Controller and PID Controller
As can be seen from Fig 8, fuzzy PID controller’s optimize performance is significantly
better than conventional PID controller’s in the control of BLDC control system. It
indicates that the control system using the former method has a faster response, a better
ability to inhibit the larger deviations. At the same time, the transition time is shorter,
and the controller parameters have better selectivity.
In summary, compared with the conventional PID control, Fuzzy PID controller achieve
a fast speed tracking control of brushless DC motor in the case of variable speed,
improving responsiveness and immunity of the overall system.
5. Conclusions
Brushless DC motor is actually a multivariable, strong coupling time varying nonlinear
system. The traditional PID control is only an approximate control within a certain
range, with low accuracy and limited performance. It is only adapted to cases which have
low control accuracy and performance requirements. To optimize motor performance,
we need to improve motor control accuracy further. According to the characteristics
of brushless DC motor, this article combine both advantages of fuzzy control and PID
control technology and design a brushless DC motor fuzzy PID control strategy. Using
MBD method, we can quickly build a fuzzy PID system simulation model, and accomplish
efficiently the system test, simulation and implementation. The simulation results show
the feasibility of fuzzy PID controller for brushless DC motor speed control system and
improvement in its performance, meeting the high accuracy requirements of brushless
DC motor. At the same time, this strategy is easy to adjust the policy improvements for
on-line control of different conditions, with good promotional value.
Acknowledgments
We would like to thank the members of the project team for the great efforts of scientific
research projects. Moreover, we greatly appreciate the reviewers’ comments that lead to
236 RISTI, N.º E5, 07/2016

an improved presentation of the results. This work was supported by Research Program
of science and technology at Universities of Inner Mongolia Autonomous Region
(NJZZ14288).
References
Informação, (16), 108–128.
Guo, Q., Zhao, X. (2008). Brushless DC motor Principle and Application. China Electric
Power Press, 14–16.
Hemati, N., Thorp, J. (1990). Robust Nonlinear Control of Brushless DC Motors for
Direct-Drive Robotic Applications. IEEE Transactions on Industrial Electronics,
37, 498–501.
Junhyuk, C., Chang, P. (2004). Development and control of BLDC motors using fuzzy
models. Proceedings of Robotics, Automation and Mechatronics, Singapore, 1,
1180–1185.
Krishnan, R. (2009). Permanent Magnet Synchronous and Brushless DC Motor Drives.
CRC Presss, Florida, 20–29.
Luo, W., Liu, J. (2009). A New Approach to the Modeling and Simulation of Brushless
DC Motors Based on Fuzzy Control. Computer Engineering & Science, 31, 147–149.
Li, Y., JIANG, L. (2016). Research on torque ripple suppression method for BLDCM
speed control system. Modern Electronics Technique, 7, 142–144.
Wang, X., Li, Q. (2003). Analytical calculation of air-gap magnetic field distribution and
instantaneous characteristics of brushless dc motors. IEEE Trans. Energy Convers,
18, 424–432.
Xia, C. (2009). Brushless DC motor control system. Science Press, 50–55.
Xu, W., Cao, J. (2011). Fuzzy PID Control of Brushless DC Motor. Science and Technology
Information, 7, 9–10.
RISTI, N.º E5, 07/2016 237

Research on the Mobile Internet Fragmentation

Economic based on Support Vector Machine
Hong Jiang*, Hong Zhao
* honganmy@126.com
College of Economics, Tianjin Polytechnic University, Tianjin 300387, China

Pages: 238–248
Abstract: The modern life is more busy and tense in mobile internet era, People
like to make full use of many scattered time during waiting, dining. And in these
fragments of time, people’s demand for mobile and Internet products has increased
and thus promoting the Internet fragmentation economy. In this paper, the
authors analyze the mobile internet user behavior preference based on support
vector machine method and discrete choice model. The result shows that the main
factors affecting the user’s mobile Internet access as network speed, price; Wi-Fi
convenience and data plan. Also we make classification of mobile Internet user
behavior preferences using support vector machines, and put forward suggestions
on the development of mobile Internet according to the relevant user preferences.
Keywords: Support vector machine, data mining, internet fragmentation,
user behavior
1. Introduction
In the mobile Internet era, modern life more busy, people will wait, meals and other
many scattered time to be fully utilized, and these pieces of time in the application of
mobile Internet products gave birth to the Internet debris in the economy. China has
527 million mobile phone users, with large user populations in the development of
Internet debris in the economy; and the mobile Internet applications in rising stage,
and has huge potential for development of Internet debris in the economy, fragments
of the Internet economy development will also become the mobile era under the new
economic growth point (Tim, 2008; Anna, 2015). Fragmentation of intent to complete
things broken into many zero block, it is described in the current context of China’s
social communication an image of the argument, that is using the fragmentation of time,
namely, the completion of each application in a very short time, incomplete and start
each time is random(Ronan, 2014; Delgado, 2015; Jiménez, 2015). And in the mobile
era, the powerful function of the Internet will be the fragmentation of time completely
covered, more people in the on the way to work, dining, leisure, wait for the fragments
of time using the network, the fragments of time has been more and more products are
now flooding the Internet network, and thus push the Internet debris in the economy
also play its endless potential, has brought immeasurable business value.
238 RISTI, N.º E5, 07/2016

Figure 1 – Internet fragmentation economy

In the traditional mode, people of information access limited to newspapers, television
and other traditional media, and limited most of these traditional media forms of
communication, such as television requires people to watch the news in a specific time,
specific space, and access to information; newspaper is every day has fixed the issuing
time, unable to meet the sudden information timeliness. The application of Internet
technology in the mobile era has completely broken the mode, it has no limit to the
mode of information dissemination, people can get information at any time, any space
through the network to get information (Georg, 2013; Eleonora, 2014). At the same
time the Internet has brought huge amounts of information also allows people to higher
selectivity of information, attention is more dispersed, is the access to information more
widely, based on the amount of information the larger, and the information acquisition
time less tolerant, may be the last just use to eat, sit car fragments of time to get the
information (Stefan, 2014; Nils, 2015). In busy modern life, people’s working time not
only to work 8 hours, in order to adapt to the rhythm of the high, high competition,
high efficiency of modern society, people must constantly use his spare time learning,
to enrich themselves, must be for the family to family effort, pay time, thus truly
considered leisure time will be more limited, more fragmented, showing a characteristics
of “fragmentation”. And in the fragments of time enhance mutual exchanges between
friends, through the network is undoubtedly the best way, blog, micro blogging, micro
channel, QQ and other series can be the application of social software in the fragments of
time has become a necessity of modern life, which is the Internet fragments in modern
life in an important manifestation of economic.
2. Support vector machine method and discrete choice model

In the analysis of mobile Internet user behavior preference, under the background
of big data, data mining is popular, and after a lot of literature proof, compared the
commonly used data mining methods in Bayesian classifier, decision tree and neural
network algorithm and support vector machine algorithm, accuracy than support
vector machine algorithm to get high, and the key is to support vector machines on the
generalization ability has a huge advantage, also has many other advantages, especially
in solving nonlinear and multidimensional pattern recognition and its advantages
RISTI, N.º E5, 07/2016 239

Research on the Mobile Internet Fragmentation Economic based on Support Vector Machine
more prominent. And through the previous chapter analysis of the characteristics of
the mobile Internet users, give full consideration to the fragments of mobile Internet
user behavior analysis data in the level of performance is nonlinear and complexity in
data analysis level performance for high dimension cross, based on using support to the
amount of machine method can give full play to the nonlinear and high dimensional
pattern recognition in the performance of many unique advantages, but also conducive
to future mobile Internet user behavior analysis on the promotion.
Figure 2 – Support vector machine
2.1. Support vector machine method

From the beginning, the support vector machine is proposed to classify the. First by
seeking the optimal classification hyper plane, support vector machine (SVM) was
originally developed for binary classification problems, the basic idea is to find a can the
different samples were separated from the optimal classification hyper plane, so that
the generalization ability of the best. In order to make the empirical risk and confidence
interval are minimal, and ultimately make the actual risk to meet the requirements.
Linearly separable support vector machine, some of the literature is also called hard
margin support to airsickness, we trained in the correct partition set in the correlation
plane, according to the method of interval, to obtain the optimal hyper plane.
Figure 3 – Optimal hyper plane
240 RISTI, N.º E5, 07/2016

The optimal plan is represented by the mathematical model:
1 1
min ϕ ( w ) = w = (w ⋅ w) (1)
2
2 2
Constraint conditions are:
yi (( w ⋅ x ) + b) − 1 ≥ 0 (2)
The original problem is transformed into dual problem, and the Lagrange function is
introduced to solve this problem.
∑α ( y ( ( w ⋅ x ) + b ) − 1 )
m
1
L ( w, b, a ) =
2
w − i i
(3)
2 i =1
Firstly, the partial derivative of Lagrange’s function on W and B can be derived from the
extreme condition.
∇b L ( w, b, a ) = 0 (4)
∇ a L ( w, b, a ) = 0 (5)
Can be concluded:
m
∑α y
i =1
i i =0 (6)
w= ∑α y x
i =1
i i i (7)
The dual problem for the original optimization problem is:

m m m
1
max a −
2
∑∑ ai a j yi y j xi ⋅ x j + ( ) ∑a j (8)
i =1 j =1 j =1
m
s.t . ∑a y ( x ⋅ x )
i =1
i i i j
(9)
The solution of the original problem is obtained according to the following calculation:
RISTI, N.º E5, 07/2016 241

m
w* = ∑a y x
i =1
i i i
(10)
m
b* = y j − ∑a y ( x ⋅ x )
i =1
*
i i i j (11)
2.2. Discrete choice model

In the classical econometric model, the explanatory variables are usually assumed
to be continuous variables. However, when we analyze and model the data, we can
meet the explanatory variable Y is not a continuous variable, but a discrete variable.
Because the variable is discrete, this kind of regression model is called the discrete
choice model. Because the number of variables and for the values of variables, discrete
choice model has several situations are as follows: (1) if for variable value only two,
known as the binary choice model, including repeated observations can get the binary
choice model and repeat observation value and will not be able to get the binary choice
model. (2) If the dependent variable is more than two, it is a multi-choice model,
including a general multiple choice model and a sort of multiple choice models. In real
life, when in the face of economic to choose, then there may have a variety of options
for the, people may be confused in the end is selected metal or real estate investment;
the same in the choice of the employer, this confusion, people do not know to choose
private or state-owned enterprises; the same in the selection of a communication
network also have a variety of options, whether to choose China Unicom and mobile
telecommunications is also are a kind of choice.
Thus, the discrete choice model is mainly refers to the individual in decision making,
from a series of options to choose a model for their own, when these options are
not is random, subject to the following points: first in quantity cannot be unlimited,
second option must be in set of options to find. Finally, an option is mutually exclusive,
individual to choose an option can no longer go select other options. At the beginning,
the discrete choice model is only used in traffic demand, and later with the development
of the need, the method has been involved in various industries. By the random utility
theory can be seen that consumers make decisions are based on the principle of the
maximum utility of the utility, so the probability of J product selection for consumers to
use the following formula:
(
P = Pr Vnj + ε nj > Vnk + ε nk ) (12)
ε is assumed to be out of touch in the MNL model, and its distribution is in accordance
with a class of extreme value. The distribution function of the density function and the
application of the distribution function are shown in the following two formulas:
( ) ( )
f ε nj = exp −ε nj exp − exp −ε nj ( ( )) (13)
242 RISTI, N.º E5, 07/2016

The probability that the j product is selected by the consumer n can be calculated
according to the following formula:
Pnj =
(
exp V ε nj )
J (14)
∑
k =1
exp (V ε nk )
3. Mobile Internet user behavior preference

3.1. Data collection
In this paper, the most intuitive user contact user interface User Interface (UI) were
studied by the interface of the operating habits of users to mobile Internet user behavior
analysis of preference for the user interface (UI) research on user behavior preference,
must through the application on the terminal of mobile Internet the user terminal
application variety, Baidu, NetEase, Tencent news client is representative and largest
of several other domestic news client, news client mode are similar, and the news
browsing in the text and pictures is an important part of the mobile phone users of
mobile Internet traffic consumption, so the research on Baidu, NetEase. The actual data
of Tencent news client based on mobile Internet environment the user about the news
reading model, first we respectively All the news of the show (PV) and the amount of
clicks (Dick) conducted a statistical analysis, found the following rules. For the amount
of news show to do statistical analysis, the basic statistical distribution table is as table1.
The table show the number in less than 10 about 77%, show in less than 20 accounted
for about 88%, perform the higher number of 50 news proportion only 0.0011, in which
the overall show the number, minimum 1, maximum value of 650, the mean of 11.1, with
a median as 4.
Cumulative Cumulative
Show Show
number news distribution number news distribution
times times
rate rate
1 30531 0.2054 0.2054 11 2868 0.0193 0.7893
2 20113 0.1353 0.3408 12 2390 0.0160 0.8054
3 15659 0.1053 0.4462 13 2141 0.0144 0.8198
4 12294 0.0827 0.5289 14 1905 0.0128 0.8326
5 9742 0.0655 0.5945 15 1633 0.0109 0.8436
6 7662 0.0515 0.6461 16 1440 0.0096 0.8533
7 6229 0.0419 0.6880 17 1256 0.0084 0.8618
8 4866 0.0327 0.7207 18 1174 0.0079 0.8697
9 4033 0.0271 0.7479 19 1095 0.0073 0.8770
10 3289 0.0221 0.7700 20 1006 0.0067 0.8838
Table 1 – Statistical analysis of news
RISTI, N.º E5, 07/2016 243

3.2. Mobile Internet user segmentation

The premise of the subdivision of the user is to select the dimension of user segmentation,
this dimension is determined according to the actual target of data mining projects.
In mining the data to determine the success, not all the data can predict the ideal
results, data quality is good or bad directly determines the success of the model, only
to meet the requirements of the quality of the data to get the ultimate success of the
model. When determining the integrity of the project indicators, the index system of
the mobile Internet is very important, it directly affects the project indicators. To meet
the requirements of the variables selected out, and then produced into an analysis
of the wide table variables, the wide table can be included into all the variables, this
time, ready to work basically completed, the next step is to carry out data mining. The
establishment of user index system directly affects the success of data mining.
business Proportion
Appstore, Android 20.3%
Micro-blog; 25.6%
QQ, MSN, fetion; 50.3%
WeChat、 27%
Happy, everyone, on campus 10.3%
SkyDrive 5.7%
Mobile reading 26.7%
Online game; 16.7%
Listen to or download music online 29.7%
Watch or download videos online 18.3%
buy goods online 15%
Send and receive mail 16.7%
Search or browse information 34.7%
Table 2 – Mobile Internet service statistics

From the table can be seen mobile phone Internet users access to most business QQ,
Fetion business, and search and browse information is more also. Users are using WIFI
mobile Internet WIFI Internet access is now an important feature of the mobile Internet,
the characteristics of different traditional mobile phone users, so we regard WIFI as a
research aspect
business Proportion
Often used; 21%
Occasionally use 29%
No, the phone does not have the WIFI function; 9.3%
No, the phone has a traffic package 6.3%
No, there’s no hot spots around WIFI. 19.7%
Table 3 – WIFI mobile Internet Statistics
244 RISTI, N.º E5, 07/2016

4. Internet user behavior classification

With the increasingly fierce competition in the mobile Internet market, how to make
customers satisfied and let users more for the mobile Internet pay has become the
mobile Internet business, especially the core task of telecom operators, at the time of
purchase, the needs and desires of consumers determines they buy, enterprise value,
but by offering consumers, different user for the enterprise to provide the value also
differ. According to statistical analysis, we know, most of the profits in enterprises is
caused by the small number of people. That is to say most of the customers are unable
to create profit for enterprise’s, and even let enterprises in the loss of the state, so to the
enterprise, find out the small part for enterprises to create the customer’s profit is very
important. When it comes to carry on the classification to the user the, but because too
many external factors considered in the classification, and the internal relations between
these factors is also very difficult to figure out. At the same time, for the number of users
increasing. We identify the six significant factor affect the user is willing to increase
mobile Internet access charges, three of which are relevant, three negative correlation,
users think that Internet speed is more important that the CMCC use cumbersome and
expensive the positive influence on the mobile Internet users are willing to increase
the cost of the mobile Internet will, think the phone traffic packages will be enough,
mobile Internet service for the social networking site (Kaixin, Renren, school, etc.), non-
internet related businesses the negative to affect the willingness of mobile Internet users
are willing to mobile Internet payment. We put the six main dimensions as we support
vector machine of main dimensions, such as follow-up research predicts the relationship
process, saving the cost of the investigation, and at the same time, can improve the
success rate and efficiency of classification research and the six dimensions can also
label free quantification in the operator’s background, not through direct investigation
of user classification prediction, under the background of big data, to more accurately
grasp the user’s individual needs. Users value the phone Internet speed; remember
for user as A. valued win hotspots using convenience; remember to user as B. for WiFi
hot price sensitivity as C. users think enough phone traffic packages; remember for D.
mobile phone users access to business for social networking sites, as E. mobile phone
user’s access to business for non-internet related business, recorded as F.
A B C D E F category
0 0 0 0 1 0 0
0 1 1 0 0 1 1
1 1 0 0 0 0 1
0 0 1 0 0 1 0
0 0 0 1 0 1 0
0 1 0 1 0 0 0
0 1 0 1 0 0 0
0 0 0 0 1 1 0
0 1 1 0 0 1 0
Table 4 – Part of the training sample
RISTI, N.º E5, 07/2016 245

In this paper the experimental part is implementation of packet libsvm software SVM
based on the, libsvm developed by National Taiwan University researchers, as a software
program, in pattern recognition and regression is simple, quick and effective. Also the
program and a great advantages can not only performed in the windows operating
system, but also can be implemented in the Linux operating system, at the same time,
the procedure of transplantation with good performance, through the pledge of relevant
code of the software provided by the modified can achieve its transplantation. Set known
training set:
T= {( x , y ),( x , y )( x , y ) ∈ ( X × Y ) }
1 1 2 2 i i
m
(15)
Select the appropriate kernel function K and the appropriate parameter C to construct
and solve the optimization problem:
m m m
1
max
2
∑∑a a y y K ( x ⋅ x ) − ∑a
i =1 j =1
i j i j i j
j =1
j (16)
Select a* of a less than positive classification, and calculate:
m
b* = y j − ∑a y ( x ⋅ x )
i =1
*
i i i j
(17)
The decision function is constructed as:
 m 

 i =1
∑
f ( x ) = sgn  ai yi K ( x ⋅ xi ) + b* 

(18)
For different kernel functions in SVM, we compare the prediction classification accuracy
of the test set, and the parameters are selected on the segment to get the parameter
C = 2, g = 0.1, as shown in table:
kernel function Classification accuracy Svmtrainparameter options

Radial Basis Function 98.6667%(148/150) c = 2, g = 0.l
linear 95.3333%(143/150) c = 2, g = 0.1
sigmoid 58.6667%(88/150) c = 2, g = 0.1
polynomial 98.6667%(148/150) c = 2, g = 0.1
Table 5 – Comparison of different kernel functions

We use and Rand C and Rand G the two procedures of randomly generated parameters
c and G, and the use of this set of parameters to train the SVM, finally burst prediction
and classification accuracy is obtained, and through the parameter to obtain the optimal
classification accuracy compared, as shown in the following table, from the table can
be seen, random selection of the parameter C and G cannot guarantee the classification
accuracy, sometimes can be chosen empirically to more appropriate parameters,
246 RISTI, N.º E5, 07/2016

but there is no guarantee that this is the best parameters, even if it can make correct
classification rate reached the highest. But the research on the algorithm to find the
parameters to ensure the sense and the best classification accuracy rate is higher.
Running times Random parameter C Random parameter g Classification

accuracy
1 36.6754 89.8734 44.6667%
2 60.4562 5.2753 91.3333%
3 3.8756 79.2386 45.3333%
4 77.5109 21.8952 54.6667o/o
Table 6 – Comparison of different parameters
5. Conclusions
Mobile Internet users is always the core and key of enterprise research, but also the
future of mobile Internet companies and telecommunications operators to sustainable
development and competitive source. How to explore and understand the preferences of
the user’s behavior and to provide personalized service is a business concern and willing
to invest capital research important topic. The research in the theory of support vector
machine (SVM), discrete choice model were reviewed and summarized in this paper based
on is discussed in detail. The development of mobile Internet and user behavior research,
through in-depth analysis of the statistical learning theory and support vector machine
(SVM) of the basic theory and characteristics, will support vector machine (SVM) theory
is introduced into the mobile Internet user behavior research field, an effective solution
to the problem of empirical research in complex nonlinear and high dimensional pattern,
and to improve the generalization ability of the learning machine, breaks through the
traditional user behavior research methods and ideas.
In the discrete choice model to obtain the significant influence factors based on, using
support vector machine (SVM) to classify the mobile Internet user behavior preferences,
give full consideration to the fragments of the mobile Internet in the data analysis the level
of performance is nonlinear, mobile Internet behavior of complex of the data analysis level
performance for high taboo of cross, on the basis of this, give full play to the SVM method
in nonlinear and high dimensional pattern recognition exhibit many unique advantages.
Using MATLAB software to carry on the analysis, selection of C-SVM binary classification
machine, taking with RBF kernel function and parameter of the model and kernel
function with cross validation method for optimizing, after training the classification
correct rate satisfactorily, and comparing the different kernel functions and parameters
of the classification accuracy. The results show that the method of this study, research
shows that the support vector machine method for mobile Internet user behavior, and
use SPSS MODELER software to analyze the importance of each predictor variable, and
mobile phone traffic packages can meet the user is one of the most important prediction
variables, divided by six significant dimensions of performance finally the level of the
user is divided into four groups, which focus on the efficiency of business professionals
and white-collar workers, are sensitive to the price of a high degree of migrant workers,
service personnel, pay attention to mobile phone traffic packages is enough students, and
RISTI, N.º E5, 07/2016 247

not sensitive to price but the pursuit of the user experience of the individual boss, each
class the groups showed their unique preferences and behavior patterns, and according
to user preferences related to the development of mobile Internet is proposed.
References
Anna, D., Antonio, R. (2015). Net Neutrality and internet fragmentation: The role of
online advertising. International Journal of Industrial Organization, 43, 30–47.
Delgado, A., Velthuis, M. (2015). Proposal for a continuous improvement IT governance
framework at financial institutions. RISTI-Revista Ibérica de Sistemas e Tecnologias
Eleonora, B. (2014). The Internet of Things vision: Key features, applications and open
issues. Computer Communications, 54, 1–31.
Georg, G. (2013). Competition, regulation, and broadband access to the internet.
Telecommunications Policy, 37, 1095–1109.
Jiménez, M., Vicente, E. (2015). Safeguard selection for risk management in information
systems: a fuzzy approach. RISTI-Revista Ibérica de Sistemas e Tecnologias de
Nils, W. (2015). Qualitative indicators for smart city business models: The case of mobile
services and applications. Telecommunications Policy, 39, 218–240.
Ronan, K., Jonathan, E. (2014). Influencing online grocery innovation: Anti-choice as a
trigger for activity fragmentation and multi-tasking. Futures, 62, 155–163.
Stefan, B., Oliver, F. (2014). Surfing alone? The internet and social capital: Evidence
from an unforeseeable technological mistake. Journal of Public Economics, 117,
73–89.
Tim, S., Mei-Po, K. (2008). The Internet, mobile phone and space-time constraints.
Geoforum, 39, 1362–1377.
248 RISTI, N.º E5, 07/2016

Research on Teaching Evaluation of College Physical

Education Based on Data Mining and Network Survey
Zhengdao Lu 1, Yibo Zhao 2,*
* zyb@vip.163.com
1
Sports Institute, Lang Fang Teachers University, Langfang 065000, China
2
Sports Institute, Hebei Normal University, Shijiazhuang 050024, China
Pages: 249–259
Abstract: With the rapid development of the data mining technology, network
data analysis begins to play a significant impact on statistical analysis, especially the
performance evaluation. Balanced iterative educing and clustering using hierarchies
(BIRCH) is an integrated hierarchical clustering algorithm, its biggest characteristic
is to complete the efficient clustering of large data sets by using limited memory
resources, so that the BIRCH data mining algorithm can be applied to large data sets
and data flow. In this paper, the author put forward a new method for evaluating
the sports education by using online survey, the result shows that the new method
is more effective in training exercise habits. The main goal of physical education is
to enhance students’ physical fitness, improve sports skills and increase the spirit
of cooperation. The result shows that more than 82.4% students satisfied with the
evaluation content of new physical education, 33.8% students agree that the new
method is conducive to improve the sports skills. Through the students’ physical
quality test, the result shows that after the implementation of the new evaluation,
physical education promotes students’ interests and hobbies, and improves the
students’ enthusiasm for learning and exercise.
Keywords: BIRCH algorithm, physical education, data mining, sports skills
1. Introduction
Physical education teaching evaluation is an important link in the teaching of physical
education, physical education teaching evaluation is directly related to the basic concept
and orientation of physical education, which is related to the progress and development
of the students, and the realization of the goal of college physical education teaching.
Serious deviation in current colleges and universities sports teaching evaluation mode,
there is a singular assessment content, and assessment tools and methods monotone
problems, these problems and physical teaching evaluation ideas, methods, contents,
index and unreasonable has a close relationship (Basu, 1987; Djelassi,2013). This should
provide the correct guidance, service and incentive function of physical education
teaching evaluation to a large extent bound up the reform of the hands and feet, hinder
the implementation of quality education and the realization of the goal of physical
education. Damage to many students’ self-esteem and self-confidence, and dampen the
RISTI, N.º E5, 07/2016 249

Research on Teaching Evaluation of College Physical Education Based on Data Mining and Network Survey
enthusiasm of many students’ sports learning, hinder the progress of students’ sports
learning and development, seemingly objective, in fact, is not fair to students (Eijck,
2008; Delgado, 2015). There are also teachers’ teaching performance ability, while
ignoring the performance of their creativity and the degree of students’ participation
in physical education and sports activities. The ordinary universities sports teaching
evaluation system lag has become a constraining bottleneck of the physical education
reform and development of China’s colleges, P.E. teaching evaluation reform is
imperative under the situation. Therefore, to build a accords with the requirements of
the times really promote the development of the teachers and students in Colleges and
universities sports teaching evaluation system and has the advantages of reliability, the
corresponding evaluation index system is very necessary.
The main goal of college physical education teaching is to enhance students’ physical
fitness, improve health, improve the ability of physical education and lifelong physical
training, and cultivate the habit of lifelong physical education and physical exercise. The
traditional evaluation of students’ physical education only focuses on the physical and
sports skills, ignoring the other aspects of the development of students, using a unified
single evaluation methods and standards to evaluate all students (Varley, 2011; Li,2014).
On the evaluation of classroom teaching objective is also emphasis in “identification”
and “classification” belongs to a reward and punishment evaluation, evaluation results,
mainly used to determine whether the teacher possesses a reward or punishment
conditions, these have a direct impact on the realization of University’s sports teaching
goal (Wei, 2012). With the continuous deepening of the reform of physical education
and the renewal of educational ideas, it is imperative to reform the teaching evaluation
of physical education. Study on teaching evaluation of physical education in Colleges
and universities in China, mainly in the late 1990s, sports teaching experts and scholars
on the concept of school physical education evaluation in our country, reform target and
evaluation method were theoretical research and exploration, has made great progress.
The research on the reform of College Students’ physical education performance
evaluation is also reported, but only from different perspectives with different research
methods on the existing physical education teaching evaluation of a certain aspect of
reflection and discussion.
2. Large scale data set mining algorithm

2.1. Sampling methods
Data mining is an important part of knowledge discovery. It is found that the data
mining is done through the establishment of the mining model and the corresponding
algorithm. Nowadays, many kinds of models and algorithms of association rules,
classification, clustering, and decision tree and so on are provided in order to select and
use. Sampling is a classical statistical technique, has been studied for over a hundred
years of history, especially the random sampling technique, there are many basic
principles, and describes the effectiveness of random sampling.
In the field of data management, a small part of a subset of data sampling can capture
the basic features of the data through extraction to represent the total data set, and
250 RISTI, N.º E5, 07/2016

Figure 1 – Data mining process
according to the sample set to obtain approximate query results, or based on the data
mining of the sample set. Figure 1 is a representative sampling method of classification
in data mining. The probability is selected according to the data items are the same, the
sampling method can be divided into uniform sampling and sampling bias of two. As the
name suggests, in the uniform sampling each data item in the same probability of being
selected in the sampling bias, selected probability of different elements may be different.
2.2. BIRCH algorithm

If the original data set is very large scale that not a complete read into memory cluster,
you can first on a data set of data identification, summarizes some of the data, in order
to reduce the memory space occupied by the data, realize the clustering of large data
sets. Iterative educing and Clustering Using Hierarchies Balanced is an integrated
hierarchical clustering algorithm. Its biggest feature is the use of limited memory
resources for efficient clustering of large data sets, while minimizing the cost of I/O.
Figure 2 – CF (cluster feature) tree
RISTI, N.º E5, 07/2016 251

BIRCH algorithm firstly stores the data set in a compact compressed format (clustering
feature tree), and then directly on the compressed data structure, and does not need
to cluster the original data set. The I/O cost of the algorithm is linear with the size of
the data set. In general, the single pass scan data set can get better clustering results.
Optional follow-up of multiple scan can further improve the quality of clustering. In the
micro clustering stage, the BIRCH method is characterized by CF (Feature Clustering)
to describe a cluster, forming a CF tree. CF tree cannot be more than the amount of
information loss in the premise, compression hierarchical structure of the cluster.
The core of BIRCH method in the micro clustering stage is the construction of CF tree.
Assume that the data object is a data set of P n dimension. CF tree is a 3 dimensional
vector, which contains the information of the data objects. Its definition is as follows:
CF = ( n, LS , SS ) (1)

n
LS = ∑ xi (2)
i =1
n
SS = ∑ x 2 (3) i
i =1
The use of CF (cluster feature) can be derived from a part of Statistics:

n
∑x
i =1
i
LS
=
x0 = (4)
n n
∑( x − x0 )
2
i
i =1 nSS − 2 LS 2 + nLS
=R = (5)
n n2
n n
∑∑ ( x )
2
i − xj
=i 1 =j 1 2nSS − 2 LS 2
=d = (6)
n (n − 1) n (n − 1)

Among them, the radius of R and diameter of d are reflected closely around the centroid
of the cluster. The radius of R measurement the average distance to the centroid of the
members, the average distance of the objects in the cluster was measured by d.
SS1 SS2 2 LS1 LS2

d* = + − (7)
N1 N2 N1 N2

252 RISTI, N.º E5, 07/2016

3. Physical education evaluation system

3.1. Evaluation system construction
With the research of deep people, more and more researchers believe that the real sense
of learning is the process of learners’ active construction. Therefore, valuable learning is
considered to be the main role of the learner based, emphasizing the cognitive activities
of new knowledge , emphasize the knowledge and skills of the application, transfer,
emphasizes the process and experience of teaching methods, emphasizes the important
effect of interactive teaching in the subject of emotion, attitude and value outlook index
and. At the same time, the model of human development, to understand the relationship
between education and human development is the core of the education idea, namely
respect individual and individual differences, teaching students in accordance of their
aptitude realize the whole person education, lay the foundation for lifelong learning in
the individual. Accordingly, accordingly, new evaluation concept arises at the historic
moment. Evaluation is not to complete a task, but a continuous process; evaluation is
used to support education, it is to teach and learn the main, the essence of a component,
integrated throughout the teaching activities of each link. The basic goal of the evaluation
is to education, and promotes the student’s performance, not just to check the students’
performance; evaluation is of service learning, the purpose is to improve the efficiency
of learning, is the power and source of learning; evaluation is for people’s lifelong
development. Evaluation should pay attention to the individual situation and needs,
respect and reflect the individual differences, to stimulate the spirit of the individual, in
order to promote each individual to achieve their own value.
1. Physical education learning evaluation criteria: social development not
only require students to have a strong physique and a certain level of motor skills,
also require students to have healthy psychology and good social adaptability,
and this requirement is the traditional unified sports learning evaluation
standard can’t satisfy the. Therefore, the sports learning evaluation criteria
necessary to diversify, the diversification of the evaluation standard of the sports
study not only can improve the reliability and validity of the evaluation result
to a great extent, but also can give a good evaluation of the incentive function.
According to the requirements of the physical education curriculum teaching
guide, it can be divided into the standard of the absolute standard, the relative
standard, the standard of the individual difference. Therefore, in order to make
the evaluation can really play an effective role in promoting students’ learning,
the educational value of the body can be better realized value, the evaluation
of physical education should be based on the individual internal standard, and
combined with the absolute standard, the relative standard to evaluate.
2. Learning evaluation of physical education: traditional sports learning
evaluation model is based on students’ physical quality and the movement ability,
heavy summative evaluation, process evaluation, repressed the development of
students’ personality and hobbies of the play, not for each student to provide
different degrees of successful experience, which restricts the most of the students
physical and mental health of sustainable development, a serious impediment
to the implementation of quality education. Sports Curriculum Teaching
Instruction Summary attaches great importance to comprehensive evaluation of
RISTI, N.º E5, 07/2016 253

students’ physical learning, emphasize the development in students’ evaluation

of physical fitness and sports skills at the same time, pay more attention to
the evaluation of students’ learning attitude, emotion and ability and so on, in
order to truly embody, all student oriented education concept, truly reflects the
assessment of the fairness and the education function.
3.2. Research tools and methods

The evaluation standards and contents to the students, but also design and make the
corresponding evaluation tools. In most cases, these evaluation tools usually appear in the
form of evaluation table. Not only to reflect the students’ sports performance evaluation
table, and evaluation should also reflect the students learning process and learning
attitude scale. The student sports basic theory knowledge evaluation can take oral, written
examination, knowledge contests and other forms, according to the assessment results are
converted into the score of the student’s basic theoretical knowledge of sports learning. In
order to simplify the examination procedures and workload, the students physical fitness
test and score can be combined with school each year to implement the “Student Physique
Healthy Standard” to, the student sports study physical scores and with reference to the
health standard of students’ physique three physical quality score conversion into a. The
choice of assessment content according to the nature and characteristics of sports skills
assessment project, adopt the method of quantitative or qualitative results make the
corresponding evaluation table. Physical and motor skills progress evaluation method,
first of all, according to the student physical fitness and sports skills raw scores and test
scores, to measurement result is subtracted from the original scores for the progress and to
progress points control table compiled by physical and motor skills improved evaluation
physical and motor skills progress performance assessment.
By using different evaluation and measurement methods, can collect different types of
data and evidence. Commonly used methods are: standardized test, standard reference
test, based on the results of the evaluation, student work sample, the performance of the
students observation, survey and interview. Teachers need to collect the data and evidence
for analysis, the formation of a student’s physical learning situation analysis results, and
the objective description of the current learning situation of students. To pay attention
to the following questions in the analysis: coping in the appropriate team to collect data
for analysis; response from the data of various assessment methods for comprehensive
analysis, in order to fully describe the state of development of the students. If there is a
longitudinal data, including trend analysis and if I can get the comparative data of other
groups, through comparative analysis of the development of students.
4.1. PE learning evaluation
First, we should evaluate students’ learning is a learning process and evaluation,
including physical fitness and motor skills, cognition, learning requires attitude and
behavior, communication and the spirit of cooperation, such as the performance of
the situation, in favor of the principle of operation , determined by the expert opinion
254 RISTI, N.º E5, 07/2016

to assess student learning sports more appropriate five aspects: (1) physical and (2)
motor skills (3) basic theoretical knowledge Sports (4) learning attitude (5) the spirit
of cooperation. Then apply the key features of the survey prepared by the PE learning
the five evaluation, please Colleges 40 PE teachers to express their views, and then
select the content according to the five “very important” Number Percentage (see Table
1), weight content calculated for each weight, calculated for each content selection
is to look “very important” in the five percentage content selection percentages very
important number in proportion. Followed by another four to calculate the weight of
the contents were 0.24,0.21,0.21,0.13 and give scores result: physical and motor skills
20 points; physical fitness and motor skills’ progress 30 points; 20 points physical
basic theoretical knowledge; learning attitude 20 points; show affection and spirit of
cooperation 10 points.
Physical Theoretical Learning Spirit of

Evaluation Motor skill
fitness knowledge attitude cooperation
Number 35 39 35 34 21
Percentage 87.5 97.5 87.5 85 52.5
Table 1 – Evaluation in PE learning content

Students’ modern and new sports learning evaluation content and value questionnaire
results show (see Table 2), the students of the new sports learning evaluation content
and value satisfaction rate was 82.4, and current only 35.3, suggesting that the vast
majority of students to reflect the development as the goal, pay attention to sports
learning effect and process, dilute the evaluation results of the new physical education
learning evaluation of the content of the affirmation and approval, so as to implement
the new sports learning evaluation system has laid the main foundation.
very Very
content factor satisfied General Dissatisfied
satisfied dissatisfied
Current Number 2 13 35 48 44
Evaluation percentage 1.4% 9.2% 24.6% 33.8% 30.9%
New Number 11 74 32 18 7
Evaluation percentage 7.8% 52.1% 22.5% 12% 4.9%
Table 2 – Students’ satisfaction with the current evaluation
Students’ modern and new sports learning incentive effect investigation of evaluation
method. The results show in table 3: that the new sports learning evaluation methods
form of conscious exercise habits and lifelong sports consciousness, grasp the method
and the skill, to improve physical fitness, to experience the fun of sports and work sense;
have an incentive to form good sportsmanship and teamwork spirit, identity of student
evaluation approach to new incentive effect was significantly better in the current
evaluation method.
RISTI, N.º E5, 07/2016 255

Current Evaluation New Evaluation

Factor
effective commonly invalid effective commonly invalid
Physical 18 43 47 61 14 10
fitness 12.7% 30.3% 33.1% 42.9% 9.9% 7%
35 17 20 48 20 7
Motor skill
24.6% 12.7% 14.1 33.8% 14.1% 4.9%
Theoretical 16 23 19 18 38 17
knowledge 11.5% 15.2% 13.4% 12.1% 26.7% 11.9%
Learning 13 26 41 27 45 9
attitude 9.1% 18.3% 28.9% 19% 31% 6.3%
Spirit of 16 24 18 34 18 4
cooperation 11.3% 16.9% 12.7% 24.6% 12.5% 3.9
Table 3 – Students’ recognition of the motivation
4.2. Improvement of teaching performance

Physical fitness and motor skills evaluation should first follow the progress of the case
of biological evolution “use and disuse” principle, fully reflected in the same period of
physical and motor skills of students with different sport levels the rate of progress
was generally slow down the growth of the law, that the more low levels of student
movement, the greater the level of increase, the opposite is the small, according to the
principle of progressive score, combined with the actual preparation of PE teaching
physical and motor skills rate of progress evaluation form. Reflect students’ progress
and better incentives for student rates are above 95%, while the rate applicable to the
various colleges and universities, compared with 35.5% agreed. The results show in
Table 4, the vast majority of teachers in this table design methods and ideas have a
positive attitude, but believed to be actually develop appropriate evaluation methods
and evaluation forms according to the teaching of the school.
content very good good commonly Not so good Not so good

18 16 2 2 0
Design ideas
45 41 5 5 0
Students 15 18 6 1 0
progress 37 45 15 2.5 0
8 13 17 2 0
Incentive effect
20 32.5 42.5 5 0
0 3 11 25 1
Applicability
0 7.5 27.5 62.5 2.5
Table 4 – Evaluation of physical fitness and motor skills
256 RISTI, N.º E5, 07/2016

Through physical education teachers pick him think very important indicators and sort,
more than 50% of the number of the first five indexes were chosen (Table 5): sports class
attendance; consciously take the initiative to participate in sports activities; in sports
activities to dedicate; proactive thinking to achieve the goals and repeated practice;
conscientiously accept the teacher guidance. Choice of learning attitude of the five
evaluation index from the point of view, the basic can reflect the students to participate
in sports learning attitude, to set up students’ understanding of sports and promote
students to form correct and active attitude to sport.
factor Attendance participation involvement practices instruction

Number 37 34 31 24 21
percentage 92.5 85 77.5 60 52.5
Table 5 – Evaluation of learning attitude
Students to use new sports learning evaluation method of evaluation of physical performance
and the current evaluation method of evaluation results were compared, results show in
Table 6, application of new ways to assess students sports scores of pass rate, average scores
than applying the current ways to assess student sports scores were significantly improved,
and the rate of excellent and good, excellent rate both had no significant difference. Indicating
that new ways for students with poor physical fitness and motor skills base provides through
its own efforts to pass the opportunity, overcome the horizontal movement of a high starting
point, no drawbacks need through the efforts can get high marks, better play the positive
role of learning evaluation of physical education and sports.
Content fail pass good excellent Average score
Current 13 31 71 27
77.9±8.59
Evaluation 9.2% 21.8 50 19
New 2 22 77 41
81.0±6.32
Evaluation 1.4 15.5 54.2 28.9
P value <0.01 <0.05 <0.05 <0.05 <0.01
Table 6–Comparison of PE learning evaluation 4.3. Effective of evaluation index
Students’ quality test results in Table 7, were significantly increased compared with the
implementation of new evaluation scheme for the end of the school year three quality test
scores and the beginning of the semester diagnostic test results, statistically significant
differences, the author thinks that although there are many reasons to improve the
quality of students, but the main reason is after the implementation of new sports
learning evaluation methods, physical education promotes the students’ personality and
interests play, also for each student to provide varying degrees of success experience, has
greatly promoted the students in and out of the PE curriculum of physical education and
exercise actively and consciously.
RISTI, N.º E5, 07/2016 257

project Standing long jump (m) 1000m/800m Shot-put (m)

Gender male female male female male female
Early semester 2.20±0.12 1.68±0.13 239.72±81.92 240.57±8.24 6.83±1.73 4.81±0.44
After a year 2.41±0.14 1.76±0.12 221.34±67.43 231.21±7.46 7.21±1.52 4.93±0.37
P value <0.05 <0.05 <0.05 <0.05 <0.05 <0.05
Table 7 – Student’s achievement comparison
target effective commonly invalid

16 18 0
Reflect the teaching process
40 45 0
18 19 0
Promote teachers’ progress
45 47.5 0
21 17 0
Improve learning interest
52.5 42 0
15 14 0
Improve sports skills
37.5 35 0
25 14
Interaction
62.5 35 0
Table 8 – Survey results of effectiveness of classroom teaching
Physical education teachers chosen for the study of classroom teaching evaluation of
the validity of the questionnaire survey results show that in Table 8, it is investigated
in the teacher the classroom teaching evaluation index reflected in the elements of the
teaching process, promoting the progress and development of teachers, is conducive to
the performance of the creativity of Teachers, improve the learning interest of student
sports, help to improve the students’ physical fitness and sports skills, is conducive to
the recognition of the role of interaction between teachers and students and the effective
rate was 100%, thus scientific implementation teachers’ classroom teaching evaluation
and teachers’ consciousness to participate in evaluation and lay a good foundation.
5. Conclusion
The research method on the basis of the theory of building college sports teaching
evaluation system with promote all-round development of students and teachers to
improve the concept of education evaluation, is also in line with the requirements of
“National Ordinary University Sports Curriculum Teaching Instruction Summary” and
lifelong sports goal orientation, with certain scientific and theoretical role, to adapt to
the trend of the development of College PE teaching evaluation. On the basis of the
learning evaluation of physical education and sports content, results of an empirical
study of the index evaluation contents, indicators, and the weight of the evaluation table
258 RISTI, N.º E5, 07/2016

and existing evaluation methods, compared with a certain objectivity, effectiveness and
operability, reflecting the promote all-round development of students and teachers
constantly improve the evaluation concept, in line with to promote the all-round
development of students not only pay attention to the students’ physical fitness, physical
and mental health, but also to discover and develop the potential of students, including
the development of the students in various aspects of moral character, physical education
learning desire and ability, cooperation and communication, personality and emotion,
not the athletic performance as the main means of evaluation requirements.
The establishment and implementation of the new evaluation system of physical education,
teachers’ ideas are the key to renew, we must vigorously promote the transformation of
the concept of education, the unity of people’s ideological concept. So that the evaluation
system of the development and implementation of a teacher’s conscious action, to
promote quality education, improves the quality of physical education teaching process.
Learning evaluation of physical education and sports is a service in the sports teaching,
to serve the students development must change the past that kind of attention to the
evaluation and selection approach, the evaluation activities become students self-learning
progress required of a feedback, diagnosis and incentive means. The purpose of teacher’s
classroom teaching evaluation is to promote teachers’ progress and improve the quality
of teaching. Evaluation should give full play to the main role of teachers, the evaluation
work and the teacher’s personal development plan. Teacher’s self-evaluation should be
one of the most important forms of classroom teaching evaluation.
References
Basu, K., Jones, E. (1987). The growth and decay of custom: The role of the new
institutional economics in economic history. Explorations in Economic History,
24, 1–21.
Djelassi, S. and Decoopman, I. (2013). Customers’ participation in product development
through crowd sourcing: Issues and implications. Industrial Marketing
Delgado, A., Velthuis, M. (2015).Proposal for a continuous improvement IT governance
Eijck, K., Lievens, J. (2008). Cultural omnivorousness as a combination of highbrow,
pop, and folk elements: The relation between taste patterns and attitudes concerning
social integration. Poetics, 36, 217–242.
Li, G. (2014). Destination island effects: A theoretical framework for the environmental
impact assessment of human tourism activities. Tourism Management Perspectives,
10, 11–18.
Varley, P. (2011). Ecosophy and tourism: Rethinking a mountain resort. Tourism
Management, 32, 902–911
Wei, C. (2012). Application of Multimedia–Aided Project–Teaching Mode in Cultural
Education, IERI Procedia, 2, 538–542.
RISTI, N.º E5, 07/2016 259

Research on the Competency Model of Innovative

Entrepreneurial Team based on Network
Information Data Mining Technology
Jia Liu *
*Liujiacheery@163.com
Science and engineering research institute, Harbin University of Science and Technology, Harbin 150080, China
Pages: 260–272
Abstract: Since the 20th century, with rapid development of information

technology, the enterprise competition more embodied in the product technology
competition. Product technology innovation, integration of resources, and create a
team of technical innovation has become an important way to improve enterprise
R&D. In the process of innovation, entrepreneurial team plays an important role, but
also should pay attention to cultivate entrepreneurial team competency. By using
data mining method, the author constructs the evaluation index of the Innovation
and entrepreneurial team competency. And find out three dimensional indexes
as innovation dimension (0.123), team dimension (0.557) and entrepreneurial
dimension (0.320). In the team dimension, entrepreneurial team should focus
on improving team cooperation ability; in the aspect of entrepreneurship,
entrepreneurial team should enhance relationship ability.
Keywords: Information technology, innovation ability, team competency,
entrepreneurial dimension, data mining
1. Introduction
Since twentieth Century, with the development of society and economy, the competition
of enterprises is more and more reflected in the competition of product technology.
Through product technology innovation, integration of resources, to create a team of
technical innovation, has become an important way to improve enterprise R&D capabilities
(Antonio,2014; Abreu,2015). With the gradual maturity of team theory and practice, the
technology innovation team has been accepted and applied by more and more enterprises
(María, 2012; Freixo, 2014). In the process of innovation, entrepreneurial team plays an
important role, in order to transform the results of innovation, but also should pay attention
to cultivate entrepreneurial team competency (Yuzhu, 2011; Hung, 2012). The so-called
entrepreneurial team refers to the two or more than two members of the company’s strategic
decisions have a direct impact, and to a certain extent, to enjoy or occupy the ownership of
the enterprise. To cultivate a very important point for the innovation and entrepreneurial
team is to improve the ability of innovation and achievement transformation rate. In the
process of corporate entrepreneurship, outstanding entrepreneurial team to achieve long-
term performance requirements of the ability, skills, knowledge and other characteristics,
260 RISTI, N.º E5, 07/2016

and have a fully capable of entrepreneurial task, is the entrepreneurial team competency. In
general, and the team competency should possess the following several points: encouraging
attitude to effective team performance, on behavior and skills needed to effectively perform
the tasks, as well as the team task concept, principles and relevant knowledge. And in
the aspects of learning, emotional control, commitment ability, strategy, relationship
coordination, knowledge sharing, innovative thinking, the opportunity to grasp, they have
reflected the entrepreneurial competency (Olimpia, 2014). Entrepreneurial spirit, member
heterogeneity, management collaboration innovation, technical experts, knowledge
team and other innovative entrepreneurial team features. Therefore, the entrepreneurial
team can be regarded as entrepreneurial spirit, because it can be a team collaboration
and recognition, recognition of entrepreneurial opportunities to share. And risk sharing,
collaborative spirit, collective innovation, and sharing and team entrepreneurship cognitive
level dimensions. After study the entrepreneurial team competence, in the innovation
environment, compared to the traditional team entrepreneurship, innovation and
entrepreneurial team competence is entrepreneurial team to obtain high entrepreneurial
performance should have the qualities and entrepreneurial ability, skills, and knowledge as
a whole (Luis, 2014).
The idea of innovation and entrepreneurship is constant innovation, the new business
as the cornerstone of a reasonable change of innovation, and then the correct
understanding of business opportunities, and identification, open drill business.
Innovative dimensions, dimension and team entrepreneurship dimensions constitute
the following indicators, respectively is: collective innovation capacity, technology
trend judgment, learning ability, risk sharing, cooperation ability, cognitive share.
Especially the collective innovation and technology trends judgment and ability to
learn from the expert interviews and research; Risk-sharing, cooperation ability and
cognitive share in the study of table to team up to power; and organization ability,
relationships, cooperation, strategic skills, opportunity comes from table unremitting
research to entrepreneurial competence power. In terms of dynamic development
perspective, a 3D space respectively by the three dimensions of team dimensions,
dimensions of entrepreneurship and innovation dimensions constitute, thus forming
a innovation and entrepreneurial team competence. With the development of the
enterprise day after day, taking the entrepreneurial dimension as the premise, the
birth of innovation and entrepreneurial team competence, the team dimension and
innovation dimensions maintain a positive momentum.
2. Data mining technology

Data mining is a process from a large number of incomplete, noisy, fuzzy and random real
data to extract the hidden information and knowledge. From the data mining definition
can see that data mining is an important technology in the field of knowledge discovery.
It involves artificial intelligence, machine learning, pattern recognition, statistics and
other high technology fields, specific technology including characterization, association,
clustering, and prediction analysis. Data mining in the field of Internet, mobile Internet,
telecommunications, finance, scientific research has been widely used, such as Facebook
friends recommended, recommended am AZON and Taobao goods, Bank of the anti-
fraud analysis etc.. The traditional data mining technology based on relational database
RISTI, N.º E5, 07/2016 261

Research on the Competency Model of Innovative Entrepreneurial Team based on Network Information Data Mining Technology
and data warehouse, the data were calculated, to find the hidden in the data model and
relation, and in the large scale data the calculation data access and statistics was made,
the whole data mining process need to consume a large amount of computing resources
and storage resources,
The emergence of cloud computing, making the data mining platform has a new direction
of development, but also makes a new generation of data mining platform has become
possible. Cloud computing is a computing platform that can provide dynamic resources,
virtualization and high availability. Cloud computing platforms can be used to develop
high performance applications. But for data mining, the massive data itself has the
problem of noise, heterogeneous, complex algorithm, technology, and so on, and now
the cloud computing platform does not provide data reduction and other functions. From
the industry’s understanding of cloud computing, cloud computing dynamic, scalable
computing capabilities make it possible to efficiently massive data mining. Cloud computing
SaaS function understanding and standardization, making the data mining based on
SaaS has the support of technology and theory, but also makes the data mining for the
popularization and enterprise oriented. Cloud computing platform based on data mining
services, mining algorithm parallelization, mining algorithm based on the perspective of
the construction of data mining SaaS platform. The cloud based data mining platform
structure adopts a hierarchical thought: first supporting the bottom of the cloud computing
platform, and the use of cloud computing platform provides a distributed storage and
distributed computation ability data mining computing power of parallel implementation;
secondly the data mining platform in the design uses a distributed, pluggable components
of ideas, support multi algorithm deployment and scheduling; finally data mining platform
provides ability of the algorithm to employ the services of external exposure and support
different business system calls, and achieve the business system recommended, mining
and other related functional requirement is convenient.
Figure 1 – Cloud computing

Cloud computing distributed storage and distributed computing to promote the new
generation of data mining platform for change. Figure 2 is cloud based data mining
262 RISTI, N.º E5, 07/2016

platform architecture. Considering the parallel and distributed algorithm of mining

algorithm and recommendation algorithm is a special and big topic, so the article does
not include the parallel and cloud content of the specific algorithm.
Figure 2 – Cloud based data mining platform architecture

Data mining methods used in user behavior classification are neural network, bayesian
network and decision tree method.
•• Decision tree: Decision tree is also called a classification tree or a regression
tree. It is also a machine learning method, which is developed by Quinlan.
Decision tree is utilized to approximate the discrete value function classification
and evaluation method of an inductive learning system, decision tree has the
advantages of simple structure and convenient operation, and has anti-noise
robustness and can clearly express the learning results, is currently a widely
used machine learning algorithms. But the disadvantage of the decision tree is
that there is no better way to prevent the data, and it is not very good to solve the
overlapping problem, and it is difficult to optimize.
•• Bayesian network: Bias network is a probabilistic network, which is a graphical
network based on probabilistic reasoning, and the Bias formula is the basis of
this probability network. Bayesian network is a mathematical model based on
probabilistic reasoning, called probabilistic reasoning is through some of the
variables of the information to obtain other probabilistic information, based
on a probabilistic inference in Bayesian network (BN) is proposed in order
to solve the uncertainty and incomplete, it to solve complex equipment does
not cause the uncertainty and correlation of the fault has great advantages, in
many fields obtained widely application. Bias theory shows that the subjective
belief should follow the principle of probability, the correct induction should
be based on the independent way of reasoning and spread through the Bias
network, it is the fault of the network structure of the training is too complex
•• Artificial neural network: Neural network source of interdisciplinary research
in physics, psychology and neurophysiology, Neural networks are classified
according to the network structure and learning methods, which include:
feedback network, random neural network, competitive neural network and
single layer forward network, multi layer forward network. Currently used the
most widely used neural network is BP network. It belongs to multi layer to
the network, the technology is mature and has powerful computing ability,
RISTI, N.º E5, 07/2016 263

but the complexity of the statistical model but also to artificial neural network
processing problems bring some negative effects. The disadvantages of artificial
neural network is its generalization capability is relatively weak, lack the ability
to explain, it is difficult to use explain the network each node of the decision and
method to determine whether the network is feasible, and due to the algorithm,
artificial neural network is very easy to fall into local minima.
Figure 3 – Bayesian network

Approximate linear separable support vector machines, also known as soft interval
support vector machines, is established in the case of linear separable, that allows the
existence of the wrong sub samples. Here we only consider a loss function of SVM,
the original problem is converted as:
m
1
φ ( w,ξ ) = ∑
2
w + C ξ i (1)
2 i =1
The problem is to minimize the constraints, That is:
(2)
yi (( w ⋅ x ) + b ) − 1 ≥ 0 (3)
Then carries on the derivation of the dual problem, need to introduce the original
problem of Lagrange’s function:
∑α ( y ( ( w ⋅ x ) + b ) − 1 + ξ ) (4)
m m
1
L ( w, b, a ) = ∑
2
w + C ξi − i i i
2 i =1 i =1
264 RISTI, N.º E5, 07/2016

First find the Lagrange function on the W, B, ξ Partial derivatives derived from extreme
conditions:
∇b L ( w, b, a ) = 0 (5)
∇ w L ( w, b, a ) = 0 (6)
∇ξ L ( w, b, a ) = 0 (7)
Calculated:
m
∑α y
i =1
i i = 0 (8)
m
w= ∑α y x
i =1
i i i
(9)
C − α i − β i = 0 (10)
Take into the plane equation, get the decision function:
 m 
∑
f ( x ) = sgn  ai yi ( x ⋅ xi ) + b *  (11)
 i =1 
3. Construction of index system

3.1. Index System
Based on the technological innovation and entrepreneurial team characteristics,
entrepreneurial team competence research, we believe that technological innovation
and entrepreneurial team competence is the sum in the environment of innovation of
science and Technology entrepreneurial team achieves high entrepreneurial performance
required knowledge, skills, abilities and qualities. Compared with the traditional team
entrepreneurship, science and technology innovation and entrepreneurship stressed the
continuous generation of new innovation, innovation and transformation of the new
business, and identify business opportunities to produce commercial results.
So in the entrepreneurial team qualified for the force of the dimensions of
entrepreneurship and team dimensions based on, increasing the innovation dimensions
constitute, technological innovation and entrepreneurial team competency: from
the three dimensions of the dimensions of entrepreneurship, dimension and team
innovation dimensions, a total of 10 indicators, in particular: opportunity competence,
strategic competence, ability, organization ability, cognitive sharing, cooperation ability,
RISTI, N.º E5, 07/2016 265

Figure 4 – Entrepreneurial team competency
risk sharing, learning ability, the technical trend of judgment and collective innovation
capacity. The opportunity ability, strategic capabilities and relationships, the ability
to organize main reference entrepreneurial competence table study; shared cognition,
cooperation ability and risk sharing main reference team competence table study;
learning ability, the technical trend of judgment and collective innovation can force
mainly refer to the relevant innovation research and expert interviews.
first-level indicators second-level indicators third-level indicators

Learning ability C1
Innovation dimension B1 Technology trends C2
Collective innovation capability C3
Innovation and Cognitive ability to share C4

entrepreneurial Team dimension B2 Cooperation ability C5
team competency
evaluation The risk sharing capacity of C6
A Opportunity capability C7
Strategic capability C8
Entrepreneurial dimension B3
Relation ability C9
Organizational capability C10
Table 1 – Index system

Analytic hierarchy process to determine the decision maker’s experience to quantify, a
decision-making problem is described as a hierarchical model, and then quantitative
analysis and calculation. With respect to a CGI elements, we use 1 ~ 7 scaling the expert
assignment method each element of B interlayers in pairwise comparison and judgment
to determine the intermediate layer B element of a CGI elements relative importance,
construct pairwise comparison judgement matrix A. Judgment matrix A (relative
importance of each factor relative to the scientific and technological innovation and
entrepreneurial team competence), see table 2.
266 RISTI, N.º E5, 07/2016

A B1 B2 B3
B1 1 1/4 1/3
B2 4 1 2
B3 3 1/2 1
Table 2 – Judgment matrix A
3.2. Index weight

Calculate the geometric mean of all the elements in the matrix A.
n
wi = n ∏aj =1
ij (12)
We can obtain:
( )
T
wi = w1 , w2 , w n (13)
To be normalized, that is to calculate:

n
wi = wi ∑w
j =1
i (14)
Obtain:
w = ( w1 , w2 , Kwn ) (15)
T
This is the relative weight of each element of Bi to B:
( ) = (0.123,0.557,0.320)
T T
w = w1 , w2 , w 3
Calculation of the maximum eigenvalue of theλmatrix A Max

n
( AW )i
λ max = ∑
i =1 nwi
Among them (AW) i is the first i element of vector AW:
 1 1 / 4 1 / 4   0.123  0.369 
    
AW = 4 1 2   0.557  =  1.688 
 3 1 / 2 1  0.320  0.967 
n
 AW 
λ max = ∑   = 3.01832
i =1  nwi 
RISTI, N.º E5, 07/2016 267

Judgment matrix to calculate the judgement matrix, the largest eigenvalue corresponds
to the characteristic to (weight), and then calculate the consistency checking and ratios,
if C. R < 0.1 that matrix consistency test requirements, if not satisfied, continue to
modified until can meet only.
The consistency check of the calculation results is C.I, the formula and the calculation
result is:
C ⋅ I = ( λ max − n ) ( n − 1 ) = 0.009162
Identify the corresponding R.I = 0.58 by Table 3.
n 1 2 3 4 5 6 7
R.I 0 0 0.58 0.90 1.12 1.24 1.32
Table 3 – Average random consistency index R.I

The random consistency ratio, the formula and the result are calculated:
C .R = C .I ( R.I ) = 0.009162 0.58 = 0.01579 < 0.1
According to the random consistency ratio C. R < 0.1, indicating that the consistency
of the judgment matrix A can be accepted, the middle layer B of CGI a weight value
w = (0.123, 0.557 ,0.320), t is acceptable. Similarly, compared with the middle layer
B elements, we use 1 ~ 7 scaling the expert assignment method of the elements of
the underlying C for pairwise comparison and judgment to determine the relative
importance of underlying C elements of B elements of middle layer, we construct
pairwise comparison judgment matrix B 1, B 2, B 3, B 4 can also be obtained the weight
of all elements in the third layer, and the judgment matrix consistency test, see table
4 and table 5 and table 6.
B1 C1 C2 C3 W1
C1 1 3 1/2 0.334
C2 1/3 1 1/3 0.142
C3 2 3 1 0.525
Table 4 – Calculation of weight in the second index system C1, C2, C3
B2 C4 C5 C6 W2
C4 1 1/3 2 0.252
C5 3 1 3 0.589
C6 1/2 1/3 1 0.159
Table 5 – Calculation of weight in the second index system C4, C5, C6
268 RISTI, N.º E5, 07/2016

B3 C7 C8 C9 C10 W2
C7 1 5 1/3 2 0.245
C8 1/5 1 1/7 1/2 0.063
C9 3 7 1 5 0.572
C10 1/2 2 1/5 1 0.119
Table 6 – Calculation of weight in the second index system C7, C8, C9, C10
The bottom element C for the synthetic weight goals of A, each index weight see table 7.
first-level indicators second-level indicators third-level indicators

C1 (0.334)
Innovation dimension B1
C2(0.142)
(0.123)
C3(0.525)
C4(0.252)
Innovation and Team dimension B2
entrepreneurial team C5(0.589)
(0.557)
competency evaluation A C6(0.159)
(1.000)
C7(0.245)
Entrepreneurial dimension B3 C8(0.063)

(0.320) C9(0.572)
C10(0.119)
Table 7 – Each index weight
4. Innovation and entrepreneurial team competency

fuzzy evaluation
Fuzzy evaluation method arises from the existence of fuzzy phenomena. In real life, the
deterministic and random phenomena are things were more common in the two kind
of way of existence, the fuzzy phenomenon is between two common ways, it is difficult
to of the precise and quantitative description of the phenomenon, a comprehensive
evaluation method of fuzzy mathematics for us provides a measure of the tool. A fuzzy
comprehensive evaluation method is based on fuzzy mathematical basis, the application
of fuzzy mathematics in the “fuzzy comprehensive evaluation” to carry out safety
evaluation method, fuzzy comprehensive evaluation method according to the fuzzy
membership degree theory of the qualitative evaluation into quantitative evaluation.
In the evaluation of scientific and technological innovation and entrepreneurial
team competency assessment indicators are divided into two levels, the first level of
3 indicators, the 10 layer of the second indicators, reviews are divided into 5 levels.
According to the calculation of the analytic hierarchy process, the weight of each index
is known. This paper selects Institute of science and technology innovation experts and
outstanding entrepreneurial team members consisting of 10 evaluation group, according
to 10 bits of advice given the influence the ability of global impression of each index fuzzy
comments, see Table 8.
RISTI, N.º E5, 07/2016 269

Very
Domain Weight Good General Poor
Domain U Weight A Good
Ui Ai (V2) (V3) (V4)
(V1)
U11 0.041 0.2 0.5 0.2 0.1
Innovation
dimension 0.123 U12 0.017 0.2 0.4 0.3 0.1
U1
U13 0.065 0.4 0.5 0.1 0
U 21 0.140 0.1 0.5 0.3 0.1
Team dimension
0.557 U 22 0.328 0.2 0.5 0.1 0.2
U2
U 23 0.089 0.2 0.3 0.3 0.2
U 31 0.078 0.3 0.4 0.1 0.2
Entrepreneurial U32 0.020 0.1 0.4 0.4 0.1
dimension 0.320
U3 U 33 0.183 0.3 0.5 0.1 0.1
U 34 0.038 0.3 0.4 0.2 0.1
Table 8 – Comments of index grade in S&T entrepreneurial team
The index weight vector A of the Ui obtained from the above calculation is evaluated by
the fuzzy evaluation of the object with respect to each index.
Ri = rgh
i
k ⋅m

Fuzzy transformation is carried out as follows:
Bi = Ai ⋅ Ri = bi 1 , bi 2 , Lbim 
Calculation of fuzzy reviews relative to the first layer of each indicator U:

0.3 0.5 0.1 0.1 
 
Bi = Ai ⋅ Ri = 0.034, 0.180, 0.095  0.3 0.34 0.2 0.1 
0.3 0.5 0.5 0 
= 0.038 0.060 0.020 0.006 
Normalized calculation can be:
B1 = 0.305 0.486 0.162 0.048 
B2 = 680.175 0.486 0.182 0.175 
B31 = 0.287 0.457 0.131 0.124 
270 RISTI, N.º E5, 07/2016

According to the evaluation results we can see that membership of the index system
in three innovative dimensions, dimension and team entrepreneurship dimensions for
the evaluation of good and better respectively 0.791 and 0.643, 0.744, the evaluation is
quite reasonable. At the same time, we can see clearly, for the analysis of technological
innovation and entrepreneurial team competence and evaluation results pointed out that
the team dimensions (0.557) and entrepreneurial dimensions (0.320) should be very
prominent, in specific three-dimensional refinement indicators, collaboration (0.328),
relationship ability (0.183) and collective innovation capacity (0.065) in technological
innovation and entrepreneurial team competence has an important role.
5. Conclusions
Through the analysis of the evaluation results of science and technology innovation
and entrepreneurial team competency, entrepreneurial team in addition to focus on
their own overall competency upgrade, the specific can start from three aspects of team
dimension, entrepreneurship and innovation dimensions. The ability to cooperate is
based on the cooperation of team members, team members are complementary, and the
ability to work on the basis of trust. Teamwork ability helps to enhance team innovation
ability. Enhance team collaboration capabilities to each team members take the initiative
to understand the positive qualities of the other members hold a tolerant attitude,
personal ability to fully integrate into the team, to enhance the team’s cohesion, sharing
of resources, improve the team cooperation ability. Relationship ability refers to the
ability to interact among team members, social organization and memory environment,
including the ability to establish and coordinate with the surrounding environment with
the help of social relations, communication and interpersonal communication. Ability to
more emphasis on the relationship between team members and external environment,
for technological innovation and entrepreneurial team, each member of the team find
opportunities and capital operation, social resources and the accumulation of experience
and the relationship between entrepreneurial team valuable resource in organization
in that team members to through the accumulation of past personal experiences and
resources, in Entrepreneurship and development process, by continuous with team
members to share, exchange and the establishment of new organizational resource to
upgrade team ability.
In the entrepreneurial team to grasp the technical trend judgment, play collective
innovation capacity for the sustainable development of enterprises is particularly critical,
such as the executive team discuss new product research and development, the use of
the collective wisdom of perfect market development programs, the collective decision-
making access to resources scheme etc., which will meet the requirements of scientific
and technological innovation in knowledge innovation, technology innovation and
management innovation three aspects, to technological innovation and entrepreneurial
team competent force plays a crucial role.
Acknowledgments
The work of this paper is supported by the humanities and social sciences research
project of Heilongjiang provincial education department in 2013(12532065).
RISTI, N.º E5, 07/2016 271

References
Antonio, L. (2014). From potential absorptive capacity to innovation outcomes in
project teams: The conditional mediating role of the realized absorptive capacity
in a relational learning context. International Journal of Project Management, 32,
894–907.
Informação, (16), 108–128.
Hung, Ts., Ja-Shen,C. (2012). The influence of interfirm codevelopment competency on
e-service innovation. Information & Management, 49, 177–189.
Luis, E., Vila, P. (2014). Innovation at the workplace: Do professional competencies
matter? Journal of Business Research, 67, 752–757.
María, T. (2012). Technological distinctive competencies and organizational learning:
Effects on organizational innovation to improve firm performance. Journal of
Engineering and Technology Management, 29, 331–357.
Olimpia, C. (2014). Customer Orientation, Innovation Competencies, and Firm
Performance: A Proposed Conceptual Model. Procedia - Social and Behavioral
Sciences, 148, 16–23.
Yuzhu, L., Ming,Y. (2011). The role of team problem solving competency in information
system development projects. International Journal of Project Management, 29,
911–922.
272 RISTI, N.º E5, 07/2016

Study on the Coordinated Control System of Urban

Traffic Trunk Line based on Particle Swarm Optimization
Chengtao Cao1,*, Feng Cui1, Lunhui Xu2
* Jncct@163.com
1
Guangdong Communication Polytechnic, Guangzhou 510650, China
2
South China University of Technology, Guangzhou 510640, China
Pages: 273–286
Abstract: In this paper, the author research on the basic theory of traffic
coordinated control, and analyze the formation factors and characteristics of trunk
line system. We propose a delay model of trunk traffic coordinated control based on
traffic flow, and using the actual statistics of traffic data to carry on the simulation.
Experiments shows feasibility and effectiveness of the PSO algorithm in the system,
compared with the traditional delay time control, experimental data show that the
PSO algorithm can improve the traffic efficiency and reduce the time of vehicle
delay, so as to achieve the goal of optimizing the main line.
Keywords: Urban traffic trunk, particle swarm optimization, coordinated control
system, intelligent transportation
1. Introduction
In recent years, under the background of the rapid development of the national economy,
automobile industry is thriving, car in urban traffic gradually occupied the dominant
position, become residents of an important means of transport, urban trunk road car
traffic increasingly saturated, and the urban road traffic in the world has increasingly
serious congestion. Transportation and people’s lives are closely related, but also the
lifeblood of urban development, which marks the progress of urban civilization, is the key
to the rapid development of urban economy (Wen,2008; Ren, 2013). The construction
of urban transportation infrastructure behind the pace car with increase of quantity and
residents on traffic demand and this led directly to the occurrence of a variety of problems:
traffic congestion, environmental pollution, noise pollution and energy shortage, the
countries all over the world are hard to resist (Liu, 2004;Freixo, 2014). Regardless of
the degree of economic and industrial development, the western developed countries
and developing countries are facing the increasingly serious traffic problems, without
exception. In all kinds of traffic problems, the problem of road congestion is particularly
prominent, but also the world’s major cities need to solve the primary problem (Jiang,
2011; Li, 2011). Traffic congestion and chaotic management resulting in greatly reduced
the traffic capacity of the whole traffic system, driving the vehicle time delay increases,
the running speed cannot be increased, leading to the increase of people’s travel time
cost, in a certain extent, driver’s mood impulse, go against the traffic regulations and
RISTI, N.º E5, 07/2016 273

Study on the Coordinated Control System of Urban Traffic Trunk Line based on Particle Swarm Optimization
the idea that traffic accidents occur frequently. The occurrence of traffic accidents
and further make traffic jams worse. Our country happen traffic accident incidence
although has obviously decreased, but total number of accidents is still huge, but also
resulted in the shocking casualties and huge economic losses. The problem of traffic
safety is a major threat to the safety of life and property, as well as a stable social order.
Therefore, it is urgent to improve the safety of road traffic.
In response to the traffic problems, there are always a variety of measures and solutions
are proposed, but the world is not really solving the problem. Before the most direct
way is to accelerate the pace of traffic infrastructure construction, mainly in the form
of increased quantity of road, increase the width of the road, overpass bridge erection,
optimization rules of road traffic etc., and expand the channels of traffic flow, evacuation
traffic flow, but due to the limit of funds and the city space, this method is often not
ideal (Beasley, 1993). Another way is from traffic management starts with modern,
scientific management method and traffic regulations on traffic flow effectively control,
coordination and command, to maximize the release of the capacity of the existing
road, crossing. UTCS (Traffic Control System Urban) as the core part of the intelligent
transportation system is an effective way to improve urban traffic conditions. System
in essence is the operation of the automatic coordination and control of area traffic
signal lamp, make the traffic flow more reasonable and balanced, reducing travel
time of vehicle delay and number of stops, can reduce environmental pollution at the
same time, so that he can make full use of the efficiency of the road traffic system. In
the urban traffic network, the main traffic load from the main line to bear. Therefore,
research on urban trunk line traffic flow control strategy, set up the trunk of intelligent
transportation system (ITS), and reduce the traffic delay and number of stops is very
important significance to improve the status of urban traffic congestion.
2. Model of coordinated control system for traffic trunk line

2.1. Traffic trunk system coordination control
In the staggered urban traffic network, the main line traffic is the lifeblood of the
city, the main line of traffic efficiency directly affects the traffic situation in the area
of communication. In the existing road network, although the intersection has taken
corresponding traffic signal control strategy, but due to the intersection between
the road distance is too short and the vast majority of single point control, enabled the
team to cannot be timely evacuation, leading to the road network transport efficiency
is very low, and fuel waste and serious environmental pollution is serious and so on a
series of urgent problem. Therefore, it is necessary to route traffic travelling with certain
continuity. To reduce stop time of the vehicles traveling on the trunk road sections and
intersections, ensure unimpeded trunk line, usually in the same trunk line on some
adjacent intersection traffic signal as a whole, according to the traffic flow and the
actual road conditions of coordinated control, that the lines of communication system
of coordinated control. Coordinated control system through the trunk, referred to as
the control system is mainly used to improve traffic efficiency, city trunk line traffic.
Online control system, the intersection with the signal cycle of the same, green signal
274 RISTI, N.º E5, 07/2016

ratio should be determined based on the cross export the direction of traffic flow ratio
and coordination control intersections in the green light to start time to maintain a
reasonable phase difference. When entering trunk roads of vehicles in the intersection
traffic light through keep system design speed driving, the theory can guarantee not
to encounter a red light, make in the trunk line always get the right of passage and
unimpeded, and traffic signal lights formed a green wave band. The adjustment phase
difference control method to set the green time is called green wave control, as shown
in figure 1.
Figure 1 – The formation of the green wave band

Suppose that the main traffic load in the trunk line is more than the traffic load, and the
traffic flow in the direction of the traffic flow is the main traffic flow in the main traffic
flow, and the traffic flow in the north south direction is the main traffic flow on the main
road. The function of the signal lamp is to distribute the traffic time of the intersection,
to avoid or reduce the delay caused by the traffic conflict. For trunk roads, if in the upper
reaches of the intersection in the green light after the team to saturation flow all pulled
out of, reach in the downstream intersection just is green, then the vehicle without the
need for parking and straight past. Online control system, the introduction of several
concepts: uplink and downlink, internal import and import external, phase coordination
and non-coordination phase, green belt and the bandwidth, speed.
1. Up and down: The traffic in There is an exchange of calls, used to distinguish
between two different directions is upward and downward. Research hypotheses
by wire system is east-west direction of the trunk, then can be set from west to
east flow direction for the uplink direction, from east to west road vehicle is
downward direction.
2. Internal import and external import: Internal import road is to import
directly connected with the remote control system of main trunk road; external
entrance is from trunk to trunk road entrance. After a lot of data and tests show
that the internal import road traffic on the road will be affected by the impact
of adjacent intersections and interrelated, and the external entrance road traffic
arrival is random.
RISTI, N.º E5, 07/2016 275

Figure 2 – Internal and external inlet tract in trunk line system
3. The coordination of phase and non-coordinated phase: Steer by

wire system for each signal coordination control phase is coordination phase,
it is by wire system is mainly controlled by phase, so it usually selects the first
phase for the coordination phase of the system, by arterial coordination on the
intersection between the phase difference, let the vehicle from red hinder the
smooth passage of each intersection. Then the other phase corresponding to
the coordination phase is set to the non-coordinated phase.
4. Bandwidth: Bandwidth is also called access zone, green wave duration is in
the trunk. On the main line of the intersection of unequally spaced, and uplink
and downlink need two-way keep unimpeded, so green wave band of the time
could not equal to the single intersection of effective green time. According to
the direction of travel, and the bandwidth of uplink and downlink.
5. Tape speed: The tape speed is the speed; the vehicle control system can meet
online to ensure the theoretical speed of green band. Determine the speed of the
method is as follows: a road traffic management department under the speed,
this is the management of the subjective desire, but in actual operation is difficult
to assurance; traffic flow detection is the natural rate, thus determined the belt
speed can adjust, and can reflect the trunk road vehicle actual traffic conditions.
And the bandwidth is similar, with the speed of the uplink and downlink with
speed of the points.
Figure 3 – Trunk system time distance
276 RISTI, N.º E5, 07/2016

2.2. The establishment of traffic flow model

Online control system, the coordination control section of the intersection of unified set
the size of the signal cycle, and their intersection of green signal ratio according to the
actual road traffic information data set, in order to vehicles traveling on the coordination
of trunk roads not encounter a red light, at the intersection of formulation and speed to
adapt to the phase difference. Assuming that the trunk is composed of I intersection, the
distance between the intersections is L. In the arterial coordination phase direction on
the number of lanes and lane width is proportional to the, intersection, downlink traffic
statistics of the vehicle flow were recorded for qup, qdown. In the main line of the system
uplink and downlink direction, the average speed is designed for VI, up, VI, down. Unified
system for the C cycle route, two neighboring intersections between the phase difference
of Φi, i+1, green signal intersection is the ratio λ. In the traffic line, the traffic flow of the
exit road of each intersection is the intersection of traffic flow in the direction of entrance
of the intersection: they are left turn, right turn and straight flow, as shown in Figure 4.
Figure 4 – Exit road traffic flow in trunk line system

It can be seen that the traffic flow between the intersections of the main trunk road is
interrelated with each other, so we need to consider the calculation of the traffic flow.
In this, the overall upward direction of the vehicle flow is qup, the downstream direction
of the total traffic flow is qdown. In the calculation of the total traffic flow upstream and
downstream, you can take the maximum value of their respective directions.
{ }
qup = max q1,up , q2,up ,qn ,up , (1)
qdown = max {q1,down , q2,down ,qn ,down ,} (2)
Amplification of an intersection I, in the upstream inlet direction can be seen the traffic is
straight from west to east of the traffic, as shown in Figure 5 shows the intersection traffic
diagram. Here to note the intersection of I uplink direction of imports of road traffic:
qi ,up = qi ,ws + qi , NL + qi , SR (3)
In the same way, the traffic flow in the inlet channel of the i direction of the intersection is:
qi ,down = qi , ES + qi , SL + qi , NR (4)
RISTI, N.º E5, 07/2016 277

Figure 5 – Schematic diagram of the upstream and downstream direction of the intersection
Therefore, in order to be able to express the traffic flow of the intersection of the
intersection of the main lines, the traffic flow of the intersection of each intersection can
be obtained. In this paper, a concept of the steering ratio is proposed, which is based on
the direction of the intersection i:
 i qi ,WL
 ρ L ,up =
 qi ,WL + qi ,WS + qi ,WR
 qi ,WR
 i
 ρ R ,up = (5)
 qi ,WL + qi ,WS + qi ,WR
 qi ,WS
 ρ Si ,up =
 qi ,WL + qi ,WS + qi ,WR
In the same way, turn left, straight and turn right turn in the direction of the intersection
i downward direction, then the steering ratio formula is:
 i qi , EL
 ρ L ,down =
 qi , ER + qi , ES + qi , EL
 qi , ER
 i
 ρ R ,down = (6)
 q i , ER + qi , ES + qi , EL
 qi , ES
 ρ Si ,down =
 qi , ER + qi , ES + qi , EL
3. The establishment of delay model

3.1. Team all blocked
Assuming two adjacent intersections I and i+1, the team from i to i+1, if the team just at
the green light of the i+1 completely across the intersection, then the delay was 0, that is,
the team has not blocked. Instead, if the team did not during the green light from i + 1
completely passing intersection said the team blocked while travelling team blocked and
can be divided into two kinds: the first is team arrived at the intersection of i + 1, the first
278 RISTI, N.º E5, 07/2016

car encountered a red light stop making progress. As a result, the whole team is blocked
completely; the second is part of the fleet vehicles through the intersection i+ 1 when the
lights turn red, the team partially blocked.
The traffic of the main line system is divided into the upstream and downstream
direction, the upstream direction of the traffic flow modeling. Intersection i to Ii + 1 the
road distance li, i+1, upstream direction convoy was traveling at an average speed of Vup,
from the team’s first car encountered red to red light over time tW, up, and the phase of
intersection i+1 to i recorded as Φi +1,i . Then the analysis is available:
 li ,i +1 
  mod ( C ) + t w ,up = Φ i +1,i (7)
 vup 
That is
 li ,i +1 
t w ,up = Φ i +1,i −   mod ( C ) (8)
 vup 
In the team at the intersection of green time capacity for u, in a periodic system C in
red light length tred, green long for tgreen. When intersection signal lights turn green
evacuation of cumulative vehicle used by the time set for the Tgo, up, and reach the green
time of vehicle through the intersection i+1 without time delay.
( )
qi +1,up t w ,up + t go,up = t go,up ⋅ µi +1 (9)
qi +1,up ⋅ t w ,up
t go,up = (10)
µi +1 − qi +1,up
Figure 6 – All blocked delays
The triangle area is delta ABC coordination team all blocked up direction phase delay, i.e.:
RISTI, N.º E5, 07/2016 279

1 qi +1,up ⋅ ui +1 ⋅ t w2 ,up
S∆ABC = d i +1
(
= ⋅ t w ,up ⋅ qi +1,up t w ,up + t go,up = )
all ,up
2 (
2 ui +1 − qi +1,up )
 l   (11)
qi +1,up ⋅ ui +1 Φ i +1,i −  i ,i +1  mod ( C ) 
  vup  
=
(
2 ui +1 − qi +1,up )
3.2. Partial disruption of the team
A part of the vehicle to catch up with the traffic lights, and the remaining part of the team
is blocked, it is in the red light to reach the intersection of i+1 vehicles to wait for the rest
of the red light, which is called the local team blocked.
Thus available:
(
qi +1,up ti +1,red + t go )
′ ,up = t go
′ ,up ⋅ ui +1 (12)
qi +1,up ⋅ ti +1,red
′ ,up =
t go (13)
ui +1 − qi +1,up
Figure 7 – Partial delay of the team

Coordination of the direction of the phase in the direction of the team in the direction of
the delay of the local team:
1
i +1
S∆ABC = d part ,up =
2
( )
⋅ qi +1,up ⋅ ti +1,red ti +1,red + t go,up
C ( 1 − λ )2  q (14)
⋅u
=  i +1
 i +1,up i +1
(
2 ui +1 − qi +1,up )
280 RISTI, N.º E5, 07/2016

4. Road traffic simulation

Due to traffic management is still a lack of modern high-tech means and technology,
lead managers to work hard and tense and not timely, real-time understanding of the
relevant road mutation status and control scheme of the formulated in use has serious
limitations, if urban transportation for the lack of a definite purpose and effectiveness
of grooming, will enable the traffic efficiency of urban traffic system is severely reduced,
resulting in environmental pollution and energy waste problem, in a certain extent,
reducing the efficiency of urban development, contrary to the principle of sustainable
development. The current allocation strategy is still to take the intersection of multi
time control, according to the current traffic flow can not react to the old data points
in time to develop timing scheme. The section information of each intersection is
shown in the following table 1.
Factor A road B road C road D road

Cycle /s 110 120 105 110
Line width /m 35 31 31 24.5
The intersection crosswalk width /m 3.5 3.5 3.5 3.5
Minimum phase time /s 35.14 31.25 30.95 27.98
The distance between the intersection /m 765 765 525 975
Table 1 – Experimental section data information

Intersection for example to get the daily traffic flow change curve as shown in figure 8:
Figure 8 – Intersection daily flow curve
The change of traffic flow of intersection can indirectly reflect the changes of main
traffic flow. It can be seen from the figure, every night 22:00 to the next day 5:00traffic,
belonging to the night of the traffic flow of low peak stage; from 6:00 to traffic increased,
7:00 to 9:00 is to work in the morning peak, a marked increase in traffic; day, belonging
to the normal traffic flat peak time; every afternoon of 17 :00 to 19:00 traffic larger than
a day between peak even coming close to the saturated state, a second peak in a day.
RISTI, N.º E5, 07/2016 281

As night fell the vehicle from a 20:00 decreased significantly. Based on particle swarm
optimization algorithm to optimize the trunk, according to the following steps:
1. Determine the experiment to B Road intersection as the key intersection, the
calculation of the length of the trunk cycle is 120s.
2. Set up the parameter values, population size, maximum number of iterations,
and so on.
3. According to the actual survey data input to the various data.
4. Using the improved particle swarm optimization (PSO) algorithm to optimize
the system delay model and record the experimental data.
5. Set the weights of different optimization objectives, the simulation experiment
and record the data.
6. The experimental data are collected ten times, and compared with the time
delay of the current timing control mode,
According to the data of four intersection traffic flow in the experiment section, the
control strategy is formulated and the improved particle swarm optimization algorithm
is used to optimize the control strategy.
factor A intersection B intersection C intersection D intersection

Cycle /s 112 112 112 112
Coordinated phase 72 60 58 52
green light time /s
Phase difference /s 60 56 56 79
Average speed of 44
upward direction /
(km/h)
Down direction mean 31
speed / (km/h)
The total delay 2.215
control system /h
Average delay per 1.514
vehicle /s
Average delay per 2.691
vehicle /s
Table 2 – A one-way green wave control scheme

The experimental results show, under the condition of the same sections, the minimum
delay control scheme can effectively reduce time delay and ensure that a few in the green
wave bandwidth layman to intersection vehicle can with small time delay through the
trunk roads, and effectively improve the traffic efficiency. The improved particle swarm
optimization algorithm takes into account the needs of real-time control, shorten the
time of the algorithm optimization, speed up the convergence rate, and the iteration
curve of the BHPW-PSO algorithm is shown in figure 9.
282 RISTI, N.º E5, 07/2016

Figure 9 – BHPW-PSO algorithm convergence curve

According to the obtained optimization algorithm is the intersection of green signal ratio
and phase difference data, completed by wire system with program from the drawing, as
shown in Figure 10.
Figure 10 – One-way green wave coordinated phase distance graph

As shown in figure 10, the trunk road vehicles from the A Road intersection to D Road
intersection to keep the speed of the vehicle cannot be intercepted by the red light,
continuous crossing the road intersection. But in this scheme, considering only the
uplink direction green wave effect, it will lead to downstream vehicles encountered a red
light to delay. At the same time, despite the non-coordination phase of car discharge is
far less than the coordination phase, but if in the optimization is not taken into account,
then the non-coordination phase generated by the time delay will seriously affect the
steer by wire system control effect. Coordinated phase and non-coordinated phase traffic
flow are considered, and the coordination of the phase of the vehicle delay is accounted
for the total weight of the delay of the 3/4, the non-coordinated phase of the total delay
of the 1/4 value of the vehicle delay.
RISTI, N.º E5, 07/2016 283

factor A intersection B intersection C intersection D intersection
Cycle /s 120 120 120 120
Coordinated phase
62 52 56 48
green light time /s
Phase difference /s 64 42 42 118
Average speed of
upward direction / 42
(km/h)
Down direction mean

42
speed / (km/h)
The total delay control

1.813
system /h
Average delay per

1.239
vehicle /s
Average delay per

2.691
vehicle /s
Table 3 – Bidirectional green wave control scheme
Figure 11 – BHPW-PSO algorithm convergence curve
Relative to the non-coordinated phase traffic coordination phase speaking is relatively

small, most in the green time are available through, such as two-way green wave
coordination phase situation figure 12 shows, trunk roads in both uplink or downlink
direction have formed the green wave band, the vehicle can be unimpeded on the road,
in spite of bidirectional green wave band length and little, but consider non coordination
phase delay access effect in the whole trunk system is still evident than the traditional
fixed time control scheme.
284 RISTI, N.º E5, 07/2016

Figure 12 – Bidirectional green wave coordinated phase distance graph

With this scheme, the 10 cycles of the continuous section of the road are simulated and
compared with the time delay caused by the traditional timing control, as shown in
table 4. Through the experiments fully proved the feasibility and effectiveness of the
system in the control BHPW-PSO online algorithm. Compared with the traditional fixed
time control delay time, the experimental data description algorithm BHPW-PSO can
improve traffic efficiency of traffic trunk lines, the decrease in the system of vehicle delay
time, up to the purpose of optimizing the trunk.
cycle 1 2 3 4 5
Timing control delay /h 3.936 4.098 4.175 4.302 4.699
Particle swarm optimization after delay /h 1.813 1.846 1.879 1.802 1.856
cycle 6 7 8 9 10
Timing control delay /h 4.882 4.998 5.514 6.051 6.406
Particle swarm optimization after delay /h 2.075 2.216 2.357 2.402 2.437
Table 4 – Simulation of 10 cycle delay contrast
5. Conclusions
Analyzing system trunk form factors and characteristics, delay model based on traffic
trunk traffic coordination control, and the actual statistical traffic data simulation,
develop more effective than the traditional fixed time control strategy is proposed for
traffic coordination control of the basic theory of. Firstly, a traffic model traffic system,
set line control system is divided into phase coordination and non-coordination phase.
In order to more accurately represent the traffic flow of each intersection on the main
line, the intersection of each intersection is combined with the outlet channel, and the
concept of the steering ratio is put forward. According to the determined uplink and
downlink traffic, it is concluded that the main system phase coordination of upstream
and downstream direction turn left, go straight and right transfer flow. The coordination
phase is divided into two directions of the uplink and downlink, the two direction of
the vehicle traveling at the intersection of the red light, the team will be divided into
RISTI, N.º E5, 07/2016 285

two types of all blocked and partially blocked. According to the differences in traffic,
the introduction of variable coefficient established total delay model to present one-way
green wave control delay model, two-way green wave control delay model, trunk line
delay model and route delay model different control state.
Acknowledgments
The work of this paper is supported by the national Spark Program project
(NO.2015GA780024); Program of excellent young teachers in higher education
institutions of Guangdong (NO.Yq2013180);Brand specialty construction project of
Higher Vocational Education in Guangdong Province (NO.2016gzpp044).
References
Beasley, D. (1993). A Sequential Niche Technique for Multimodal Function Optimization.
Evolutionary Computation, 3, 101–125.
Jiang, X. (2011). A multi objective particle swarm optimization algorithm based on fuzzy
learning subgroup. Computer application research, 12, 4492–4494.
Liu, C. (2004). Optimization of traffic signal coordinated control of urban road network.
Journal of Tongji University, 32 (8), 63–66.
Li, L., Zhu, K. (2011). An improved fuzzy C- means algorithm based on genetic algorithm
and particle swarm optimization algorithm. Journal of systems management, 6,
728–733.
Ren, G., Huang, Z. (2013). An integrated model for evacuation routing and traffic
signal optimization with background demand uncertainty. Journal of Advanced
Transportation, 47(1), 4–27.
Wen, W. (2008). A dynamic and automatic traffic light control expert system for solving
the road congestion problem. Expert Systems with Applications, 34, 2370–2381.
286 RISTI, N.º E5, 07/2016

Extraction Process Optimization of Total Flavones

from Moriga Oleifera Leaves Based on Computer-
aided Response Surface Method
Yi Zheng1,*, Bin Wu2, Jianmei Deng1, Shaojuan Jiang1
* zyasl@sina.com
1
School of Biological and Chemical Engineering, Panzhihua University, Panzhihua 617000, China
2
School of Resources and Environmental Engineering, Panzhihua University, China
Pages: 287–298
Abstract: Moringa oleifera contains many active ingredients, and flavonoid is

one of important effective components of moringa. In this paper, the authors put
forward extraction process optimization of total flavones from moriga oleifera
leaves based on computer-aided response surface method. The radial basis function
model shows a high degree of high dimensional nonlinear approximation, and
its precision is between PRS and Kriging. By using response surface method, the
result shows that the extraction time had a great influence on the yield of total
flavonoids. The optimum extraction conditions as ultrasonic power 300 W, ethanol
concentration as 70% (v/v), solid to liquid ratio 1:27 (g/ml), extraction time as 46
minutes, extraction temperature of 50 °C, under these conditions, total flavonoids
have rate is (47.66±0.32) mg RE/g.
Keywords: Response surface method, radial basis function, moriga oleifera
leaves, total flavones, ethanol concentration
1. Introduction
Moringa oleifera Lam belonging to the family Moringaceae Moringa genus, widely
distributed in Asia, Africa, tropical and subtropical regions. Moringa contains rich
protein, vitamins, amino acid, fat, carbohydrate, root, stem, leaf, fruit, bark; seeds are
often used to treat a variety of diseases. It is also known as the “miracle tree”. Moringa
contains many active ingredients, have a variety of physiological activities, and has
a wide range of applications and prospects in the pharmaceutical and healthcare
industries (Arti,2009; Freixo,2014; Zhang,2014). In recent years, Moringa in China,
Yunnan, Guangdong, Fujian and Taiwan, get the extensive introduction and cultivation,
its pharmacological activity also caused extensive attention of experts and scholars.
Moringa oleifera whole plant can utilization, many kinds of nutrients, rich in vitamin A,
B, C, e and calcium, potassium, iron, etc. mineral nutrition elements(Brand,1995; Ping,
2007; Ashwell, 2014). It is reported that the calcium and protein levels were 4 times and
2 times the milk, potassium is 3 times the banana, iron is 3 times the spinach, vitamin C
is 7 times the citrus, vitamin A is 4 times the carrot. In addition, it also contains a variety
of amino acids and trace elements necessary for the human body (Lae, 2014; Pereira,
RISTI, N.º E5, 07/2016 287

2015). Its nutritional value can be compared with the modern nutrition experts called
micro treasure house of human nutrition of spiraling.
•• Highly nutritious vegetable: In India, the fresh leaves as a vegetable, like
spinach leaves, can be made into soup or salad. Tender pods are also edible,
can be cooked as boiled beans, the suitable harvest time is fiber does not appear
broken pods when. Cooking seeds, you must first cook a few minutes, remove
the bitter taste of the kind of shell, and then put the seeds out on the edible seed.
Resin is a thickening of the trunk of the seasoning material, there is similar to
the effect of corn flour.
•• Plant growth promoting factor: Obtained from the leaves of Moringa
oleifera extracts of foliar application can promote healthy plant growth,
resistance to pests and diseases, to promote the results, increasing fruit and high
yield. Moringa oleifera leaf juice as plant growth promoter, can make a variety
of crops (onion, bell pepper, soy, corn, coffee, tea, hot pepper, watermelon, etc.)
25-30% yield. Foliar spray can be combined with other fertilization, irrigation
and other activities carried out
•• High quality feed: Nutritional properties of Moringa is excellent, very
suitable for livestock feed. Experiments in Nicaragua show that the biomass of
120t dry matter per hectare per year can be achieved. Moringa as a livestock
feed that is economical and high yield, which is especially important for small
livestock husbandry. BIOMASA conducted extensive experiments of Moringa as
feed for feeding cattle, Moringa oleifera (beef and dairy cows, pigs and poultry).
When the 40%-50% feed of, the milk yield of the cows and the increasing weight
of the beef cattle are increased by 30%; the average birth weight of the local
Jersey cattle was 22kg, increased by 3 - 5kg
•• Moringa seed oil: Moringa seeds contain abundant oil and extract of Moringa
oil yellow bright containing 76% monounsaturated fatty acids, mainly oleic acid.
In addition, the cold pressing of Moringa oil also contains the natural antioxidant,
which makes Moringa oil is very stable, not easy to corruption characteristic.
After degumming process of Moringa oil, the induction period is olive oil 9-16
times; even without the degumming process, inducing 4.4 times and olive oil.
After refining the Moringa oil almost no taste, stability, resistance to repeated
frying, is a kind of advanced cooking oil. In the food industry, Moringa oil can
be used as food safety non-toxic lubricating oil
Flavonoids (Flavonoid) are widely distributed in the plant community, and are important
secondary metabolites of plants, they are chromone derivatives. Mainly includes: the flavone,
flavonol, isoflavones, flavanones, check ear ketones, isoflavanone, biflavone class. Most of
the flavonoids in natural plants exist in the form of glycosides, and the few exist in the form
of free form. Flavonoids is a crystalline solid, a few (such as flavone glycosides) amorphous
powder, color of flavonoids and molecular whether there cross conjugated system and
auxochrome group (- Oh, - OCH3) category, number and position of substitution on.
Flavonoids, for example, the chromone, part of the original colorless; but in the 2 position of
introducing benzene ring, that is, the formation of cross conjugated system, and by electron
transfer, rearrangement, the conjugated chain extended thus showing color. But the study
also found that in this large number of flavonoids due to the different structure, some
showed biological activity, and some did not have biological activity, but also because of their
288 RISTI, N.º E5, 07/2016

structural differences in biological activity. So the extracting and separating compounds

with higher bioactivity of flavonoids in medicine and food industry is very important.
Currently for Moringa other nutrients there are a lot of reports, and total flavonoids of
Moringa oleifera is one of the effective components of Moringa oleifera, determination of
the content of total flavonoids in leaves of Moringa oleifera reported by ultrasonic extraction
and ethanol refluxing method, measurement results are 5.31% and 3.95%.
2. RBF response surface model

2.1. RBF response surface
Radial basis function model in high dimensional nonlinear higher order approximation with
outstanding performance, the precision between PRs and Kriging, but RBF model than the
Kriging model is easier to construct and an equal precision required valuation point less
than the number of PRs, but the linearity and low dimensional problems in effect is not ideal.
In order to make up this shortcoming, at the same time, in order to make the structure of the
radial function positive definite, there is a unique solution, the general use of RBF in the form
of RBF-LP to construct the response surface. The radial basis function interpolation process
can be reduced to solving a linear equation group, when the amount of data is significantly
increased; it is a large-scale matrix inversion problem. The solution of linear equations can
be solved directly by solving the equations directly. The main result of convergence is that the
convergence rate is similar to that of the continuous order of the radial basis function. The
results show that the computational complexity is inversely proportional to the continuity
of the radial basis function. At the same time Schaback pointed out that can’t get a faster
convergence speed and better stability of radial basis function interpolation, which is also
called the calculation complexity and the convergence speed of the uncertainty principle, the
function space can be expressed as:
N
FΦ = ∑ λ Φ ( X − X ) (1)
i =1
i i
This problem is equivalent to solving a linear equation if the matrix is used:
Aλ = f (2)
The radial basis function model is constructed:

( )
1. Select appropriate radial basis functions Φ X − X i , this is determined by the
characteristics of the implicit characteristic of the sampling point and the target
function of the unknown system.
2. M sampling points are selected in the parameter space, and the source function
values are obtained.
( )
3. The data points into the expression Aij = Φ X i − X j , So the m*m order matrix
4. A is constructed, and the linear equations Aλ = f are obtained.
T
5. The linear equations can be solved λ = λ1 , λ2 ,λm  , and then The RBF model
can be determined.
RISTI, N.º E5, 07/2016 289

The radial basis function is better for the higher order nonlinear response, and the
effect of the linear response is poor. In order to make up for this shortcoming, while
guaranteeing the uniqueness of the interpolation, it is usually used in the form of
RBF-LP, that is to add a linear polynomial:
N p
Φ(x) = ∑ λ ϕ ( x − x ) + ∑ c g ( x ) (3)
i =1
i i
j =1
j j
At this point, the available (n+p) equation, its matrix form:
 A G  λ   f 
 T    =   (4)
G 0 c  0
2.2. RBF response surface comparison test

By comparing the RBF response surface model with several other types of response
surface model, the advantages and disadvantages are summarized. The RBF
response surface and RBF-LP response surface are both random Latin hypercube
sampling, due to the random nature of the sampling point, each test results are
slightly different, but do not affect the overall results trend. The root mean square
error (RMSE) is the size of the response surface structure, and the construction time
reflects the difficulty of the response surface model in different situations.
sampling point PRS Kriging RBF RBF-LP SVR MARS

RMSE 2.8315 2.0565 3.3206 3.154 1.2464 6.9080
20
Time s 0.0940 0.0174 0.0018 0.0015 0.1 0.01
RMSE 2.2999 0.6256 1.301 0.881 0.7281 4.6987
50
Time s 0.1090 0.0395 0.0083 0.0083 0.2 1.44
RMSE 2.2976 0.0151 0.6079 0.4477 0.2069 4.5347
100
Time s 0.1570 0.2722 0.0237 0.0241 1.6 22.33
RMSE 2.2783 0.0163 0.1665 0.2581 0.1278 0.6801
200
Time s 0.1870 0.3982 0.0993 0.0992 10.3 68.68
RMSE 2.2602 1.8267e-4 0.0062 0.0056 failure 0.2254
2000
Time s 0.2030 48.335 81.706 79.0984 failure 1681.48
Table 1 – Approximate AF function (D=2) of different response surface

The test results can be seen that the Kriging model can achieve a higher accuracy when the
data points are less than the data points because of the prediction function. At the same
time, it also leads to the large amount of data that need to be stored in the construction
process. Because of the particularity of the construction method, the SVR model has
no obvious change in the high dimension, but it is closely related to the number of the
sampling points. MARS has its advantages in some functions, but due to the complexity
290 RISTI, N.º E5, 07/2016

of the regression of the spline curve, the MARS response surface is constructed slowly.
RBF response surface model based on radial basis function interpolation function can
replace the high dimension and higher order nonlinear problems. The precision is
between PRS and Kriging, and it is less time consuming and easy to calculate. It is easier
to establish than Kriging model. Has greatly improved according to the comparison
of the test results, the RBF response surface disadvantage is not suitable for linear
problems and low dimensional problems, the defects in the linear polynomial, but in
other nonlinear problems RBF-LP model and RBF model have no difference.
Figure 1 – Comparison between RBF and RBF-LP response surface model in AF function
3. Experimental method
At present, the extraction methods of flavonoids mainly include solvent method, ultrasonic
method and microwave method. The traditional solvent extraction is time-consuming
and inefficient, and the long time heating will damage the structure of flavonoids, so that
the biological activity of antioxidant activity decreased significantly. Ultrasonic method
is the use of ultrasonic crushing, cavitation, stirring and other functions, to promote the
effective composition of flavonoids, such as dissolution, with high efficiency, economic
and other advantages. The experiment with ethanol concentration, solid-liquid ratio,
extraction time and temperature as factors, total flavonoids have rate and ORAC as
response value, of ultrasonic assisted extraction process parameters optimization; by
polyamide resin to flavonoids crude extract was purified, and compare its purification
changes before and after the content of flavonoids and antioxidant activity and provide
theoretical knowledge and utilization of the development of efficient extraction of
flavonoids in the leaves of Moringa oleifera and antioxidants.
Extraction process of total flavones from leaves of Moringa oleifera: Moringa oleifera
stem leaves→crushing (60 mesh) → ultrasonic extraction (300 W), → filtration,
centrifugation (5000R/min,15min) →rotary evaporation concentration → vacuum
freeze drying→flavonoids crude extract.
RISTI, N.º E5, 07/2016 291

With the yield of total flavonoids and the value of ORAC as the index, the ultrasonic
power was 300 W, and the ethanol concentration, solid-liquid ratio, extraction time and
temperature were changed in turn:
•• Ethanol concentration: extract Moringa leaf meal several copies of each 1
g, respectively, with 50%, 60%, 70%, 80% and 90% ethanol as extraction
solvent, solid to liquid ratio of 1: 30 and 60°C under ultrasonic extraction for
60 min said.
•• Material liquid ratio: extract Moringa leaf meal in a number of copies, each 1
g, were added 5, 15, 25, 35, 45 ml of 70% ethanol and 60°C under ultrasonic
extraction for 60 minutes.
•• Extraction time: extract Moringa leaf meal several copies of each 1 g, add 25 ml of
70% ethanol, and 60°C respectively ultrasonic extractions 10, 30, 45, 60 and 90 min.
•• Extraction temperature: extract Moringa leaf meal several copies of each 1 g,
add 25 ml of 70% ethanol, respectively to 40, 50, 60, 70, 80 °C under ultrasonic
extraction for 45 minutes.
Based on the single factor experimental results, the optimal parameters of four factors
and three levels were determined by response surface analysis, as shown in Table 2.
factors
level A ethanol B material liquid C extraction time D extraction
concentration (%) ratio (g/mL) (min) temperature (c)
1 60 1:15 30 40
2 70 1:25 45 50
3 80 1:35 60 60
Table 2 – Orthogonal design

Moringa leaf meal 1 g in 100 ml flask, adding ethanol water solution, in ultrasonic
cleaner (300 W) were extracted, filtration, centrifugation, with 60% ethanol fixed
volume to 50 ml, 0.5 ml extract in 25 ml colorimetric tube, aluminum complexation
points spectrophotometric determination of total flavonoids yield unit for mg RE (rutin)
/ g, the formula is as follows:
Total flavonoids yield = ( 10 × CV1 V2 M ) × 10−3
The antioxidant activity was determined by reference to the previous experimental

methods, and the samples, Trolox and fluorescein sodium were prepared with 75
mmol/L phosphate buffer (pH7.4) for the required concentration. In 96 well plate
followed by adding 20 g L not with the concentration of the sample solution liquid and
Trolox standard solution (6.25 ~ 50 umol / L), 200 mu l fluorescent light sodium salt
soluble liquid (95.6 nmol / L), and the blank control group and control group. Set the
program of fluorescence / chemical analyzer, 37 min incubation 20 mmol/L, automatic
mixing, quickly add 20 AAPH L (119.4), automatic mixing and began to measure. The
fluorescence intensity was 485 nm, the absorption wavelength was 538 nm, and the
fluorescence intensity was 3.5 min, which was 35 times a total of times. According to
292 RISTI, N.º E5, 07/2016

the formula to calculate the area under the fluorescence quenching, the formula is as
follows:
AUC = 3.5 × 0.5 × ( f1 + f35 ) + f2 + f 3 + f 4 + + f 34 
DPPH 200 µmol / L of the working fluid is prepared with ethanol, 1 ml of sample solution
added to 3 ml of DPPH solution, shake, room temperature obstacle light reaction time
20 min. in 517 nm absorption of light values Ai, anhydrous ethanol instead of DPPH
deduction of sample solution, the bottom absorption, absorption value for Aj, with
the same volume of sample solvent instead of the sample solution as control, the
absorbance as A0. At the same time, with the same concentration of Vc as the control
group, the formula is:
Clearance rate (%) = 1- Ai − A j

 ( ) Ao  × 100

5 ml 7 mmol / L of ABTS solution and 5 ml of 2.45 mmol / l K2S2O8 solution to avoid light
reaction 12 h to ABTS + using liquid, the use of liquid a day ahead of time to prepare,
and must use the same day, before use with anhydrous ethanol dilution to ceiling light
value in 732 nm 0.70 + 0.02. Take 0.4 ml of different concentrations of sample solution
in a test tube, add 3 ml of ABTS + - the use of liquid, the room temperature to avoid
light reaction for 30 min, in 732 nm absorption value was measured a I, anhydrous
ethanol instead of ABTS + - deduction of sample solution, the bottom absorption and
absorbance values for Aj, with the same volume of sample solvent instead of the sample
solution as control, the absorbance as a 0. Put the polyamide resin into the column
(Φ6 * 80 cm). Said take 10 g flavonoids crude extract with 70% ethanol for dissolution,
wet packed column, column after the balance, first with eight times of column volume
of distilled water to water soluble impurities, and then use 10 times the column volume
of 70% ethanol elution, flow rate of 2 ml / min, 250 ml of each collected a bottle, the
completion of the elution combined eluent, vacuum rotary evaporation concentration,
freeze drying, a yellow powder, room temperature placed dryer spare.
4. Results and discussion

4.1. Single factor experimental results
A stripping solution with different polarity effect of ethanol concentration on total
flavonoids yield and ORAC values as shown in Figure 2, the ethanol concentration was
80% of total flavonoids yield reached the maximum value, the concentration of less
than 80%, the yield with increasing ethanol concentration increased with increasing
concentration of more than 80% yield decreased; ethanol concentration was less than
70% when the ORAC value with the concentration increased with increase in 70% of the
maximum value, 70% ~ 90% range decreases, this is because of the different concentration
of ethanol, with the ethanol concentration increasing fat soluble substances increase,
making flavonoids active ingredient reduced activity decreased, so select 70% as a
response surface experiment center. The range in the amount of the increase of effect of
RISTI, N.º E5, 07/2016 293

ratio of material to liquid on total flavonoids yield and ORAC value, material liquid ratio
from 1: 5 (g / ml) increased to 1: 25 (g / ml), with the material liquid ratio increases, rate
and ORAC values were significantly increased, 1: 25 (g / ml), with the material liquid
ratio continued to increase both growth slow, but the energy consumption and cost will
vary with the amount of solvent and increase, which shows that increasing the extraction
solvent can only to a certain extent improve yield and antioxidant activity, and taking
into account the cost of production and post-processing and other factors, should not
be excessive increase in the extraction solvent consumption, so the material liquid ratio
at 1: 15 ~ 1: 35 (g / ml).Extraction time on total flavonoids have rate and ORAC values
influence, ultrasonic time from 10 min extended to 45 min, rate and ORAC values were
significantly elevated, at 45 min reached maximum, 45 min and then decreased, which
may is because of the destruction of the structure with the extension of time, flavonoids
and other active substances, so select 45 min as a response to a center of the surface.
Extraction temperature on total flavonoids have rate and ORAC values influence, to rate
and ORAC value reached the maximum at 50 DEG C, 50 to 80 DEG C range, with the
increase of temperature, rate and ORAC values were significantly decreased, which may
be because of solvent viscosity of high temperature to reduce, accelerate the molecular
motion, so as to accelerate the dissolution of the biological active substances, but the
temperature is too high will makes the heat sensitive compounds degradation for small
molecule compounds, resulting in yield and ORAC values decreased.
Figure 2 – Different extraction parameters on total flavonoid yield and ORAC values
4.2. Analysis of response surface optimization

By response surface method, 4 single factor analysis of ethanol volume fraction, liquid
to material ratio, extraction time and extraction temperature were obtained. The results
294 RISTI, N.º E5, 07/2016

of the experiment and the results are shown in table 3. Through design expert software
we get the results, as shown in table 4, it shows: time (P < 0.01) and temperature
(P < 0.50) of total flavonoids extraction significantly affected by. After regression fitting,
the regression equation was obtained:
Total flavonoids yield (mg RE/g)Y1=+49.16-0.23A+0.51B+0.33C-0.38D+ 0.32AB
+0.22AC-0.75AD-0.37BC-0.10BD-0.68CD-2.18A2-1.45B 2-1.25C 2-1.97D2
ORAC(µmol TE/g) Y2=+2848.83-36.96A+42.07B+2.91C-9.22D+0.56AB-102.80AC
-58.72AD+5.81BC-14.49BD+14.70CD-235.88A2-98.97B2-151.84C2-318.20D2
Experiment Total flavonoids
A B C D ORAC (μmol TE/g)
number yield(mg RE/g)
1 1 2 2 3 45.78 2419.43
2 3 2 1 2 44.91 2557.69
3 1 2 3 2 46.36 2622.56
4 3 2 2 3 44.25 2232.37
5 1 2 1 2 46.1 2418.28
6 2 2 2 2 49.04 2877.49
7 2 2 3 3 44.55 2330.1
8 1 3 2 2 45.57 2604.24
9 1 1 2 2 45.3 2447.62
10 3 1 2 2 44.06 2360.59
11 2 2 2 2 48.53 2829.86
12 3 2 3 2 46.06 2350.77
13 3 2 2 1 46.29 2297.09
14 2 2 1 3 45.49 2343.58
15 2 2 3 1 46.94 2322.57
16 2 3 3 2 47.35 2679.68
17 3 3 2 2 45.6 2519.44
18 2 3 1 2 47.22 2606.41
19 2 2 2 2 49.49 2740.62
20 2 1 2 3 45.1 2422.15
21 2 1 3 2 46.98 2587.61
22 2 2 1 1 45.17 2394.87
23 2 2 2 2 49.95 2984.5
24 2 3 2 1 46.86 2522.59
25 2 1 1 2 45.36 2537.57
26 1 2 2 1 44.81 2249.28
27 2 3 2 3 45.92 2407.43
28 2 2 2 2 48.79 2811.69
29 2 1 2 1 45.63 2479.35
Table 3 – Design of response surface of flavonoids from Moringa oleifera leaves
RISTI, N.º E5, 07/2016 295

From table 4 we can see that the total flavonoids yield and ORAC model of the p value is
less than 0.0001, the loss of the p value is greater than 0.05, which indicates that the two
models are more significant. Through the analysis of the credibility of the two models,
the decision coefficient (R 2, respectively 0.9444 0.9323, show that model fitting degree,
good correction coefficients are respectively 0.8888 and 0.8647, 88.88% and 86.47% of
the experimental points, respectively, by the two models to explain; precision is 14.648
and 12.795, shows that the model is accurate and high degree, the model is reliable. For
the total flavonoids yield model for, a B, C, D, interaction terms ad, CD and the second
A2, B2, C2, D2 the influence were significantly (P < 0.05).
source Sum of squares df mean square F value p value Result
Model 65.14 14 4.65 16.99 <0.0001 significant
A 0.63 1 0.63 2.30 0.1515
B 3.09 1 3.09 11.29 0.0047 significant
C 1.33 1 1.33 4.85 0.0450 significant
D 1.77 1 1.77 6.47 0.0234 significant
AB 0.40 1 0.40 1.47 0.2450
AC 0.20 1 0.20 0.72 0.4094
AD 2.27 1 2.27 8.27 0.0122 significant
BC 0.56 1 0.56 2.03 0.1764
BD 0.042 1 0.042 0.15 0.7011
CD 1.84 1 1.84 6.71 0.0214 significant
A2 30.81 1 30.81 112.54 <0.0001 significant
B2 13.58 1 13.58 49.61 <0.0001 significant
C2 10.21 1 10.21 37.29 <0.0001 significant
D2 25.10 1 25.10 91.66 <0.0001 significant
residual 3.83 14 0.27
Loss of fitting 2.55 10 0.26 0.80 0.6506 Not significant
Net error 1.28 4 0.32
Total deviation 68.97 28
Table 4 – Analysis of variance for the fitted regression model

Response surface figure is the response value of the test factors A, B, D, C of the
three-dimensional space of the surface map, from the response surface analysis can be
seen on the image to see the best parameters and the interaction between the parameters.
According to the regression equation, the response surface analysis diagram and the
corresponding contour map of the different factors are shown in figure 3. We can more
intuitive to see the interaction of various factors influence on the yield of total flavonoids
in mint leaves, if the curve is steep, it indicates that the more the factors effect on the
296 RISTI, N.º E5, 07/2016

yield of total flavonoids and the corresponding performance as response value changes
the size of. The contour map can be seen there is extreme conditions should be at the
center of the circle. From Figure 3 we can see time extraction of total flavonoids ratio,
extraction temperature times, and regression analysis results also shows that extraction
time and extraction temperature corresponding to the p value of less than 0.05, reached
a significant level.
Figure 3 – Interaction effects between ethanol concentration, solid/liquid ratio, time and
temperature on the total flavonoid yield
5. Conclusions
Ultrasonic assisted extraction of peppermint leaf total flavonoids in convenient operation,
the advantages of low energy consumption, ultrasound is using cavitation technology that
certain frequency ultrasonic effect in solution, the liquid in the appropriate size of small
bubbles will happen resonance phenomenon, strong shock wave that cell wall rupture,
effective components dissolution and temperature of the system cannot be substantially
increased, does not destroy the biological active ingredients, combined with the use of
ethanol as extracting agent, non-toxic side effects, and economy, and then the response
surface method analysis that the best condition of the extraction of flavonoids. The results
can be used. The optimum extraction conditions were: ultrasonic power of 300 W, ethanol
concentration of 70% (v/v), solid to liquid ratio 1: 27 (g/ml), the extraction time was 46
minutes, extraction temperature of 50 DEG C, under these conditions, total flavonoids
RISTI, N.º E5, 07/2016 297

have rate (47.66±0.32) mg RE/g, ORAC values for (2747.17±301.51) µmol TE/g, and
predictive value (49.23 mg RT/g, 2853.99 µmol TE/g) error is 0.6% and 3.7%, indicating
that the optimization of ultrasound assisted extraction the extraction technology of
flavonoids in the leaves of Moringa oleifera has better applicability. Moringa oleifera leaf
flavonoids crude extract with strong DPPH and ABTS free radical ability and the EC 50
values were 0.45 mg / ml and 0.10 mg/ml; polyamide resin purification, the contents of
total flavonoids and ORAC were 65.89% and 5923.48 µmol TE/g, EC 50 values were 0.26
mg/ml and 0.05 mg/ml, shows that polyamide resin can effectively improve the Moringa
oleifera leaf extract flavonoid content and antioxidant activity.
Acknowledgments
The work of this paper is supported by 2016 Science and technology support program of
Sichuan Province (2016NZ0062).
References
Ashwell, N., Rofhiwa, M. (2014). Antioxidant, antimicrobial and phytochemical
variations in thirteen Moringa oleifera Lam. cultivars. Molecules, 19(7),
10480–10494.
Arti, V., Vijayakumar, M. (2009). In vitro and in vivo antioxidant properties of different
fractions of Moringa oleifera leaves. Food and Chemical Toxicology, 47(9),
2196–2201.
Brand, W., Williams, M. (1995). Use of a freeradical method to evaluate antioxidant
activity. LWT-Food Science and Technology, 28(1), 25–30.
Lae, J. (2014). Soluble extract from Moringa oleifera leaves with a new anticancer
activity. Plos One, 4(9), 1–10.
Ping, H., Chi, W. (2007). Anti-fungal activity of crude extracts and essential oil of
Moringa oleifera Lam. Bioresource Technology, 98(1), 232–236.
Zhang, W. (2014). Structural identification of compounds from Toona sinensis leaves
with antioxidant and anticancer activities. Journal of Functional Foods, 10,
427–435.
298 RISTI, N.º E5, 07/2016

Technology Innovation and Manufacturing

Upgrading Drive Mechanism based on Internet Plus
Feng Ji, Ting Zhang, Fan Wu, Bangjun Wang*
*msjifeng@126.com
School of Management, China University of Mining & Technology, Xuzhou 221116, China
Pages: 299–309
Abstract: “Internet plus” is a new format for the development of the Internet
under the innovation 2.0, Internet plus means Internet plus various traditional
industries. But this is not a simple combination; it means the depth integration
with Internet and traditional industries by using information technology and
the Internet platform, and creates new development ecology. Internet plus
manufacturing means the Internet technologies to be integrated into manufacturing
industry life cycle, so as to enhance the manufacturing industry innovation ability,
the production efficiency and resource utilization rate. Based on the panel data
analysis, the result shows that new technology has a positive role in promoting the
upgrading of manufacturing industry, the elastic coefficient of R&D investment is
0.213. Manufacturing enterprises should make full use of the advantage of network
information; develop new forms of competitive advantage industries.
Keywords: Internet plus, information technology, technology innovation,
manufacturing upgrading
1. Introduction
Manufacturing industry is the direct embodiment of the level of productivity of a
country; it is an important symbol of a country’s competitiveness, also the cornerstone
of social progress. Since the reform and opening up, with the labor, resources and other
elements have comparative advantage, the development of China’s manufacturing
industry has created a miracle, but also to support the sustained growth of China’s
economy. However, China’s manufacturing industry facing many restrictions, such as
technology innovation ability is weak, rising labor costs and need to upgrade to get rid
of the dependence on the traditional comparative advantage, competitive advantage
through technology innovation ability to improve the manufacturing sector. And our
country implement market for technology policy, hope that through the introduction
of foreign technology, on the basis of digestion and absorption of re innovation; but
due to our country enterprise R&D investment and human resources investment is
not enough, technology digestion and absorption is limited, resulting in enterprise
long-term foreign technology dependence (Tim, 2008; Ronan, 2014; Anna, 2015).
Domestic enterprise technology innovation ability is insufficient, cause our country
is in the low end of the value chain competition, restrict the development of high-
RISTI, N.º E5, 07/2016 299

Technology Innovation and Manufacturing Upgrading Drive Mechanism based on Internet Plus
end manufacturing, and the lack of power of traditional manufacturing technology

transformation, restricting the upgrading of China’s manufacturing industry. At the
same time, the development of the country firmly grasp the nuclear technology, will not
easily transfer to its competitors, the technical advantages of the developed countries,
China’s manufacturing industry to upgrade the pressure (Hai, 2014; Wang, 2015).
Internet plus manufacturing is the core of the depth of integration of the two in
the “Internet plus” era of its connotation is the Internet platform and information
communication technology into the manufacturing industry in the whole life cycle,
to improve manufacturing efficiency and resource utilization rate, innovation ability,
promote the development of intelligent manufacturing, green manufacturing and service
oriented manufacturing, and ultimately to new products, new formats and business
models and other forms of construction of modern manufacturing system (Freixo,2014;
Mandal, 2015). Internet plus has some basic elements, including technology, scene,
participants, and interactive internet protocol, has the difference, the value of Internet
is different. By agreement, for example, in the manufacturing industry scene, terminal
access in very different ways, mainly including accessed via wireless access network
(3G, 4G, GPRS, Wi-Fi), and wired access network (fiber, ADSL) corresponding to each
access different access co debate, applicable to different demand scenarios. Therefore,
the manufacturing industry is often independent of the vertical system, from all things
Internet, information aggregation, the ideal goal of intelligent applications have a
great gap. Compared with the power of the United States and Japan manufacturing
industry, China’s existing manufacturing enterprises still exist technical shortcomings,
such as relying on resource advantage to develop labor-intensive enterprises and the
high proportion of relatively low level of information, which will greatly hinder the
development of Internet plus manufacturing(Jussila, 2015). The rapid growth of the
manufacturing sector will lead to the existing information infrastructure to provide
basic services, so how to ensure the normal, enhance the enterprise information
systems to achieve coverage of basic requirements of “Internet plus” is the first
consideration. Through the study on technology innovation and upgrading Internet
plus manufacturing, path and transmission mechanism can be analyzed in the role
of technology innovation and upgrading of the manufacturing industry; and the
effect of measurable technological innovation on the upgrading of the manufacturing
industry, through the establishment of measurement model of concrete, can upgrade
the technical innovation to the manufacturing industry contribution.
2. Internet promotes the upgrading of manufacturing industry

2.1. Internet development model
The user thinking is the core of the thinking of the Internet, in the planning and design
of the product to the user participation sense in place to pass, on the need for leveraging
the Internet, such as C2B mode allows users to participate in the process of product
development. Has been due to both producers and consumers in the transaction
process of communication barriers, barriers of time and space, financial payment
barriers, the producers of the product can not completely satisfy the consumers and
the cost of product customization are very high. Therefore, in order to achieve the
low cost of production process, producers and consumers can only retreat and seek to
300 RISTI, N.º E5, 07/2016

give up individuality, which is the product of the center of the product type of small
batch of large B2C model. With the development of Internet technology, production
and marketing of both sides of communication becomes more convenient, lower cost,
express logistics unobstructed developed, financial payment means fast and diverse,
resolved the existing communication, time and space, financial payment, C2B arising
therefrom. In raw material prices continue to rise, corporate use of C2B mode, for small
and medium-sized enterprises can cut costs, and enter the virtual market on the whole to
expand market share, promoting enterprises to change their structure. For consumers,
C2B is different from the traditional e-commerce sites; it is the consumer to business
e-commerce website. C2B mode operation process: consumer demand can be released
through the website information related to demand, this will manufacturers quotation,
bidding, consumers do not have as before the real market hard to find the right vendor,
also need not labor-intensive businesses with bargaining, so just in C2B website release
information needs given a self-think reasonably can bear the price; C2B site will find
a lot of businesses have the ability to experience and to offer, bidding, consumers can
from selected high price or their favorite sellers to trade. Since businesses are involved
in bidding, it is proved that these businesses can accept the price offered by consumers.
Figure 1 – Success factors of electricity suppliers
2.2. Manufacturing the new format of the Internet

Cross border thinking of the Internet is a flat, high efficiency of thinking, that is, multi
angle, multi perspective to look at the problem and propose solutions. The main purpose
is to cross the border, by other angles of view of wisdom. Cross border thinking is the first
change in the mode of thinking. Across the boundaries of thinking is easier to innovate.
Cross-border thinking simple said is planning concept of innovation, also is the value of
grafting of other industries to the industry in an attempt to create a new brand development
and the development of the enterprise management tactics. This will allow the original
non - related and even contradictory elements of the combination of infiltration, resulting
in a new growth point. Cross border thinking has a comprehensive, often related to
RISTI, N.º E5, 07/2016 301

different industries, different cultures, different areas, the need to achieve more than
one to one of the integration of innovation. Thus requiring cross-border thinking must
be profoundly aware of different industries, different culture, different areas of category,
and so in order to achieve mastery of a variety of elements. Internet economy under the
background of cross-border thinking to burst of innovation and leading enterprises out of
the woods, each encounter bottlenecks, caught in the quagmire of the brand enterprises,
cross-border thinking, to achieve transformation, upgrading and innovation.
Under the background of Internet companies have talked about the innovation, is no
longer just before the definition of technological innovation, also includes the mode of
operation and the change of business model innovation, and form a new format industry
in these innovative change based. Success of Apple’s innovation system, Nike’s flagship
brand effect the fact that widely available on the Internet, the rapid development of
information technology, the era of rapidly changing market environment, companies
can only use innovative thinking, constantly changing to adapt to the market, to meet
consumer demand, can survive, to be bigger and stronger. The development of the
Internet and new science and technology application, the entity economy and completely
fictitious economy began to merge of blurring the boundaries of many industries, the
tentacles of Internet companies began to get involved in the manufacturing, retail,
finance, telecommunications, transportation, media, entertainment and other. Cross-
border subversion of Internet companies, in fact, is also a certain degree of natural
selection, that is, efficient integration of low efficiency. Internet companies use the
Internet industry’s network information advantages, technological advantages and
management structure to upgrade their own advantages, to accelerate the formation of
a competitive advantage of the development of new formats industry.
Figure 2 – Cross border thinking in the context of the Internet
2.3. Internet boost transformation and upgrading of the manufacturing

industry
The impact of the Internet on the manufacturing process is mainly reflected in the
process of production process reengineering. Production process reengineering, is to
302 RISTI, N.º E5, 07/2016

the whole production process of deep thinking on the exploration, subversive redesign
of the entire production process in an attempt to improve the production efficiency. The
process reengineering is to meet the market and reduce the cost. Production process
reengineering requires enterprises according to the market demand, the production
process of the organizational structure, management method, processing technology
and purchasing method of re planning the layout, in order to use the least amount of
time to provide the highest quality products at the lowest prices.
The application of Internet technology has changed the way of obtaining data and information.
In the traditional way, the data and information need to be observed by the naked eye, by hand
or by using the instrument. However, under the current conditions of Internet technology,
more new technology can be used to obtain the required data or information. For example:
shelf equipped with weight sensing device can automatically measure the number of
inventory, automatic monitoring devices to replace the duty personnel. This kind of automatic
equipment with a high degree of automation, high accuracy, 24 hours uninterrupted, real-
time data access, not affected by the environment and other advantages, help enterprises to
better implement internal control. Changed the way information is stored. Compared with
the paper medium, magnetic media or optical media with storage density, erasing without
leaving any traces of, of course, the effect of is two-way. Storage density ambassador to the
enterprise can focus on the preservation of data and information resources, to be protected,
but also makes the possibility of tampering with the data increased, and once the damage or
theft will be greater harm. Improve the efficiency of information processing. In the context
of the Internet, with the high speed of computer processing ability, can make the speed of
information processing to speed up, efficiency greatly improved. However, the impact of this
on the internal control is also a dual aspect. On the one hand, the improvement of information
processing efficiency is conducive to the implementation of more complex and effective
control measures and control methods, improve the effectiveness and efficiency of internal
control. On the other hand, with the aid of high-speed information processing capabilities,
corporate employees or the ability of the authorities to fraud will also be improved.
Figure 3 – Internet big data analysis
RISTI, N.º E5, 07/2016 303

3.1. Index selection
Comprehensive domestic scholars on the Internet technology innovation measure,
R & D personnel full time equivalent, R&D input strength and other indicators are
recognized as the evaluation index. But, considering the transformation and application
of technology innovation of manufacturing industry upgrade role focuses on technology
innovation, according to the data and research, this paper intends to choose R & D
expenditures, personnel of research and development of full time equivalent, invention
patent applications and new product sales revenue to measure the level of technological
innovation, specific study of technological innovation in manufacturing industry
upgrade, instead of just the simple use of technology innovation single indicator to
measure the. Pressure
1. R&D expenditure (RD): R & D investment is the basis for the formation
of technological innovation, its support for technological innovation is not
only to carry out experimental research, but also lies in the transformation of
technological innovation. In general, the more adequate R & D investment, the
more active the more active technology innovation, technical level is relatively
fast, and then will improve the product’s technological content and added value.
This paper intends to use the total amount of R & D expenditures to measure R
& D investment. Pressure
2. Human resources (HR): Human capital is the core of technological
innovation, and it is also the main body of the diffusion of technological
innovation and the recreation of digestion and absorption. The level of human
capital not only affects the results of technological innovation, but also affects
the application of technological innovation. The higher the level of human
capital, the higher the technological level of technological innovation, and
the conversion of the application effect is better. This paper intends to use
the full time equivalent of R & D personnel as a measure of R & D personnel
indicators. Pressure
3. Invention patent (ZL): The invention patent on behalf of an invention creation,
invention patents that enterprises to master the a core technology, monopoly
advantage can be obtained through the application of technology, bring and for
the enterprise is a high profit, thus for the enterprise provides economic resources
of technology innovation. This article selects the number of patent applications to
measure the level of patent. Pressure
4. The sales revenue of new products (PS): The new product is a product
which is produced by using the new technology and the new technology, and its
added value and technical content are relatively high compared with the existing
products. Sales of new products, on the one hand, can promote product market
update, not eliminated products to meet market demand, with the elimination
of backward products, related industries will also be adjusted. To promote
industrial upgrading; on the other hand, new technology and new process will
also with sales of products and to the diffusion and propagation, which helps to
improve the manufacturing industry, the level of technology and promote the
manufacturing industry to upgrade.
304 RISTI, N.º E5, 07/2016

In order to maintain statistical data consistency, integrity and availability, empirical

analysis will be in China’s 30 provinces and cities as a cross-sectional and collects the
2007 - 2015 the provinces of large and medium-sized manufacturing added value, fixed
assets net, labor quantity, R&D investment, invention patent applications and new
product sales revenue to the role of the empirical study of technological innovation
in manufacturing industry upgrade, to eliminate the heteroscedasticity and different
dimensions, this paper for panel data of a logarithmic scale, the descriptive statistics of
the country 30 provinces, municipalities, and the variables panel data see Table 1.
indicators LNVA LNRD LNHR LNPS LNZL LNK LNL

Mean 8.186 13.542 10.566 15.979 7.656 10.216 5.126
Median 8.262 13.662 10.744 16.110 7.595 10.626 5.096
Maximum 10.200 16.187 13.004 18.946 11.234 12.565 7.358
Minimum 5.176 9.573 6.712 11.089 1.609 4.155 2.455
Std.Dev. 1.046 1.256 1.217 1.562 1.589 1.617 1.090
Observations 270 270 270 270 270 270 270
Table 1 – Descriptive statistics of panel data

Generally speaking, the panel data model can be expressed as:
yit = ∂ it + β it X it + vit (1)
The regional development level and technological innovation activities of manufacturing

industry have obvious difference, so the panel model can produce the heterogeneity, that
is, the model parameters may be different. At the same time, in order to eliminate the
problem of multiple co linearity between the independent variables of the model, and
then to cause the regression of the model, and combined with the panel data model
selection method, we get the following model:
LnYit = c + α LnK it + β LnLit + γ Lnrdit + θ Lnhrit + δ Lnzlit + ρ Lnpsit + µit + ε it (2)
3.2. Unit root test

In order to ensure the effectiveness of the panel data model, the stability of the model is
tested before the estimation of the model. In this paper, we will use homogeneous unit
root test and heterogeneous unit root test to test the stability of each variable in two
ways, and the results are shown in table 2.
According to table 2 results of unit root test shows that, between each variable first-
order partial differential were 5% and 1% significant level in wet, that variable first-order
difference points is stable to the presence of a unit root, the variables for a first-order
single whole sequence. The endogenous nature of the variables in the model often affects
the accuracy and consistency of the empirical results. Therefore, in order to ensure the
RISTI, N.º E5, 07/2016 305

Factor LLC test IPS test ADF-Fisher test PP-Fisher test

LNVA -1.399* -0.936 69.089 20.764
∆LNVA -8.102*** -2.205** 89.9945*** 86.614**
LNRD -6.334*** -0.804 64.892 63.835·
∆ LNRD -15.624*** -5.009** 98.275*** 95.631**
LNHR -0.461 -0.586 57. 836 47.174
∆ LNHR -14. 337*** -4.552** 128.275** 104.361**·
LNZL -8.069** 2.245 51.656 90.086··
∆LNZL -15.071** -7.025** 80.877** 143.986**·
LNPS -7.185** -0.887 73.425 90.674
∆LNPS -18.634*** -7.089** 126.912** 130.577**·
LNK -2.066** -0.004 15.043 15.172
∆LNK -24.451*** -18.180** 61.639*** 71.279***
LNL -5.753** -0.772 19.04 27.908*··
∆LNL -12.655*** -2.198** 13.544** 18.886***·
Table 2 – unit root test
unbiased and consistency of the results, it is necessary to carry out the endogenous test
of variables. Through the examination of the model found that the model to estimate
the residual into the model can estimate the coefficients of the model residuals for 013
estimated p value of 856; and the fitting coefficients of the model and the parameters
estimated coefficient did not significantly change and illustrate the test pass, that is, then
the original assumptions, the model does not exist in nature. From table 3 can draw W,
the panel data model of the Hausman test p value was significantly less than 05, reject
the null hypothesis, namely that the error with the explanatory variables, not to accept
the random effects model, fixed effect model should be adopted.
Test Summary Chi-Sq.Statistic Chi-Sq.d.f Prob.

Cross-section random 66.84 6 0.0000
Table 3 – Hausman test results
3.3. Test result

According to the results of Hausman test, the panel data model should adopt the fixed effect
model. In this paper, the panel data model based on, respectively on the overall and the
eastern, central and western part of China technology innovation of manufacturing industry
upgrade the role of empirical analysis; study the heterogeneity of the regional characteristics
of technology innovation to promote the upgrading of the manufacturing industry.
The regression results from table 4 can be seen that the model is more ideal. As the models
are continuously added to relevant variables, AjustedR2 is gradually increased. Since the
model regression results and theoretical analysis, model based, to the three regional
306 RISTI, N.º E5, 07/2016

Explanatory variable Model 1 Model 2 Model 3 Model 4
0.211*** 0.207*** 0.213*** 0.212***

Lnrd
(11.876) (12.597) (13.104) (13.809)
-0.101*** -0.098*** -0.108*** -0.101***
Lnhr
(-8.524) (-8.240) (-8.490) (-8.437)
0.008*** 0.036*** 0.023* 0.007*
Lnzl
(10.543) (3.150) (1.900) (1.7%)
0.051*** 0.043*** 0.021**
Lnps
(5.903) (4.852) (2.187)
0.120*** 0.125***
Lnk
(3.039) (3.339)
0.098
Lnl
(0.596)
Adjusted R2 0.935 0.944 0.946 0.932
F test 108.523 123.499 124.934 102.809
method FE FE FE FE
Table 4 – Panel data fixed effects model
models to estimate, comparative analysis of technology innovation of manufacturing

industry upgrading the difference in effect; at the same time model results of economic
analysis and interpretation.
Model 5 Model 6 Model 7·

Explanatory variable
Eastern Region Central region Western Region· ·
0.167*** 0.221*** 0.245***
Lnrd
(6.154) (5.188) (10.588)· ·
-0.080*** -0.109** -0.187***
Lnhr
(-4.195) (-2.027) (-7.629)
0.038* 0.015 0.010
Lnzl
(1.993) (0.453) (0.542)·
0.065** 0.043** 0.022
Lnps
(2,559) (1.887) (1.591)
0.148** 0.263** 0.181**
Lnk
(3.545) (1.750) (2.811)·
Adjusted R2 0.920 0.849 0.932
F test 77.669 76.679 80.585· ·
method FE FE FE
Table 5 – Regression results of Regional Panel Data Model
4. Conclusion
From the results of regression model point of view, in addition to the R & D personnel, a
measure of technological innovation research and development investment, the invention
RISTI, N.º E5, 07/2016 307

patent applications and new product sales revenue parameters estimated coefficients
are all positive, indicating that the technology innovation of manufacturing industry
upgrade has a positive role in promoting. However, overall elements of investment scale
and elements of quality improve on the manufacturing industry upgrade of the role is
still large, indicating that China’s manufacturing industry development has not been
completely get rid of dependence on the factor inputs, the role of technology innovation
of manufacturing industry upgrade exist larger space to promote. The regression results
can be seen, for the overall sample, R & D investment, invention patent applications
and new product sales revenue of manufacturing industry to increase the value of the
elastic coefficient are positive, and significant, indicating that the improvement of
the manufacturing industry increase in value, and then to the manufacturing industry
upgrade has a positive role in promoting. The R & D investment elasticity coefficient
is 0.213, R & D investment per Zo plus one percentage point will bring manufacturing
industry increase in value to improve 0.213 percentage points; the same new product
sales revenue and invention patents every one percentage point change, caused by
manufacturing industry added value changes in the same direction of 0.021 and 0.043
percentage points. This shows that R&D investment to promote technological innovation,
and the transformation and application of technological innovation, will by improving
the manufacturing industry enterprise’s technical level, and improve the added value
of high production, promote manufacturing industry increase value to improve and
promote the upgrading of the manufacturing industry. This also proves that R & D
investment is not only for technological innovation to improve the required capital,
but also to provide relevant financial support for the transformation of technological
innovation, thereby promoting the role of the manufacturing industry to upgrade.
Also as the output of technology innovation invention patent, itself represents a kind
of advanced new technology, the upgrade of manufacturing industry role in promoting
mainly through the transformation for the new product, a new technology introduced
to new production of manufacturing industry increased value of effect. According to
the model results shows that personnel flexibility coefficient -0.108, the manufacturing
industry to upgrade the role of the obvious. This is mainly in China’s R&D personnel
accounted for the proportion of manufacturing workers is low, the total amount of about
249.3 people, accounting for only 2.5% of the total number of employees, far below the
level of developed countries. Lack of human capital, resulting in poor technological
innovation capability of our country, and is not conducive to the introduction of
technology digestion and absorption re innovation, which is also hindered the level of
technological innovation in China to upgrade, thus promoting the important reasons for
the upgrading of manufacturing industry
First the traditional manufacturing enterprises to become bigger and stronger, stand
in the fierce competition in the market does not fall, it is necessary to improve the
production efficiency, to broaden the customer base, enhance the ability of marketing
services, and actively promote the management innovation to improve the management
level. The core competitiveness of manufacturing enterprises is the innovation of
management mode. And the development of the current Internet technology makes it
possible to innovate the old business management mode of manufacturing enterprises.
The improvement of the level of the internal management of enterprises can reduce
the production cost, but also can reduce the products in the market circulation cost,
308 RISTI, N.º E5, 07/2016

and speed up the flow rate of product sales, and ultimately increase the income of the
enterprise. Secondly, to remove the inherent needs of the manufacturing industry itself,
the vast logistics platform and enormous resource advantages are proved to have the
possibility of the development of electronic commerce and network economy. First,
making the industry current development is relatively mature in the logistics network
can support the development of electronic commerce; secondly, the manufacturing
enterprises are fully aware of the various links of production and operation management
of operation and characteristics, on e-commerce platform on manufacturing industry
production and marketing system operation can be very good control. Thirdly, as the
low operating cost of electric business platform, so combination of e-commerce and
manufacturing companies can achieve lower costs, improve the overall efficiency and
thus create revenue.
Acknowledgments
The work of this paper is supported by Project of Philosophy and Social Science
Research in Colleges and Universities in Jiangsu Province (2016SJD630077); Jiangsu
province education science the 12th five-year plan project (B-b/2015/01/029); Graduate
students teaching reform project of China University of Mining & Technology (2015Y02,
2016Y03).
References
Anna, D., Antonio, R. (2015). Net Neutrality and internet fragmentation: The role of
online advertising. International Journal of Industrial Organization, 43, 30–47.
Hai, L., Tian, Y. (2014). Mobile Internet Thinking and Traditional Business Reengineering.
China Industrial Economics, 8, 15–21.
Jussila, I, Tarkiainen, A. (2015). Individual Psychological Ownership: Concepts,
Evidence, and Implications for Research in Marketing. Journal of Marketing
Theory & Practice, 23(2), 121–139.
Mandal, S. (2015). Towards an Empirical-Relational Model of Supply Chain Flexibility.
International Journal of Information Systems & Supply Chain Management, 8(3),
67–86.
Ronan, K., Jonathan, E. (2014). Influencing online grocery innovation: Anti-choice as a
trigger for activity fragmentation and multi-tasking. Futures, 62, 155–163.
Tim, S., Mei-Po, K. (2008). The Internet, mobile phone and space-time constraints.
Geoforum, 39, 1362–1377.
Wang, W., He, R. (2015). The Data Service Mode of Remote Sensing Satellite Based on
the Internet Thinking. Science Technology & Industry, 7, 42–46.
RISTI, N.º E5, 07/2016 309

Supply Chain Risk Assessment based on Support

Vector Machine
Wei Cao*, Xi Zhang
* caowei1900bj@sina.com
School of Traffic and Transportation, Beijing Jiaotong University, Beijing 100044, China
Pages: 310–322
Abstract: In recent years, with the frequent occurrence of various supply chain
risks, enterprises begin to identify the risk factors of supply chain, try to make early
warning and control of the potential risks. In this paper, the authors put forward a
new method of supply chain risk assessment by using machine learning algorithm
based on support vector machine. Based on the empirical analysis, the result shows
that macro-economic form (U11), employee innovation capability (U31), advanced
technology (U41), reliability (u42), market forecast ability (U51), on-time delivery
(U62), information transmission distortion (U81) are high risk factors that will
influence the supply chain operation. So that support vector machine evaluation
method can be effectively applied in the risk assessment of supply chain.
Keywords: Supply chain risk, lean production, support vector machine,
risk assessment
1. Introduction
Under the guidance and practice of lean production, just in time production and other
advanced theories, the relationship between enterprises is becoming more and more
closely, and the market competition is no longer the competition among enterprises,
but the competition among the supply chains (Ming, 2004; Suman, 2004). In recent
years, with the frequent occurrence of various supply chain risk, business has been
greatly damaged, more and more people are aware of the importance of supply chain
risk management, supply chain risk management has become research hotspot in the
field of supply chain management (Freixo, 2014). Supply chain risk identification and
assessment is the basis of supply chain risk management. Through the identification of
supply chain risk factors and the potential risk of early warning and control, to explore
the method for the optimization of supply chain risk management, but in practice to
establish safe and reliable and has a flexible supply chain, the supply chain enterprises
in get larger gains and risk response, enhance the competitiveness of enterprises
(Abreu, 2015; Vipul, 2004). So, we need to analyze the situation of the supply chain
management of the enterprise, through the risk recognition found the weak link in
the supply chain and its possible consequences, using scientific methods to assess the
situation and determine the supply chain risk rating (Tah,2000; Prater, 2001). Based
on agile manufacturing technology of agile supply chain is produced under this kind of
310 RISTI, N.º E5, 07/2016

background, it is enterprise from all walks of life to make full use of their own advantages,
integration of global resources, to meet the diverse needs of customers and improve the
competitiveness of enterprises in the global range indicated a new way. Agile supply
chain relative to the traditional supply chain, it can with faster speed and shorter time
to seize the market opportunity, is a management mode is very advanced, now most
of the scholars explore its advantages to study how to further reduce the operating
costs of manufacturing enterprises, how to scientific and reasonable choice of strategic
alliance partner etc., so as to better for enterprises and social services. But agile supply
chain is not a perfect management method, because the agile supply chain is a dynamic
alliance as the organizational form, with the virtual and dynamic characteristics as a
significant feature, it also generates the corresponding risk.
support vector machine (SVM) is a machine learning method, based on the statistical learning
theory of VC dimension theory and structural risk minimization principle based on the,
between the model complexity and learning ability of compromises, has strong generalization
ability and accuracy. In the small sample, nonlinear and high dimensional pattern recognition
has great advantages, overcome the neural network over learning, lack of learning, local
minimum problems, and has become the following after the research of neural network
in machine learning theory (Zhimin,2001; Jeung,2001). Therefore, it has very important
practical significance to apply the support vector machine to the risk assessment of supply
chain. Agile supply chain risk evaluation, expand the depth and scope of application of risk
management, enrich the agile supply chain management in risk identification, assessment
theory and method, to make up for the agile supply chain management theory research is
insufficient, is conducive to the construction of the system of supply chain risk management.
Agile supply chain risk assessment will improve the agile supply chain operation efficiency,
reduce the unnecessary waste and loss, help build risk prevention and control strategy in
the node enterprises, for the enterprise culture core competitiveness, expand competitive
advantage and pointed out the direction.
2. Evaluation model of support vector machine

Support vector machine in order to solve the problem of data classification at first.
The traditional neural network method can produce the learning phenomenon when the
data classification. Support vector machine (SVM) introduced an optimal hyperplane
and the training set and the hyperplane have certain distance, the distance as far afield
as can make the segmentation a blank area of the both sides of the plane reached the
maximum, so as to solve the problem of learning.
The classification surface equation is:
w⋅x +b = 0 (1)
Then it is normalized to meet the
yi ( w ⋅ xi ) + b  − 1 ≥ 0 (2)
RISTI, N.º E5, 07/2016 311

Supply Chain Risk Assessment based on Support Vector Machine
Figure 1 – SVM optimal classification line
In N dimensional space, we can assume that the data samples are in the range of the
radius of the sphere, and the function set can be satisfied.
{
f ( w, x , b ) = sgn ( w ⋅ x ) + b } (3)
SGN () is the function of the VC dimension, it is satisfied:
(
h ≤ min  R2 A2  , N + 1 ) (4)
Then, Lagrange method is used to transform the optimal classification surface into a
dual problem, and the maximum value of the following functions is solved:
n
1 n
Q (a) = ∑a i − ∑
2 i , j =1
( )
ai a j yi y j xi ⋅ x j (5)
i =1
The optimal classification function is obtained by computing:
n 
{ } ∑
f ( x ) = sgn ( w ⋅ x ) + b = sgn  ai yi xi ⋅ x j + b*  ( ) (6)
 i =1 
If the direction of the corresponding method of classification is given, the optimal
classification can be selected from these lines, as shown in Figure 2. The optimal
classification line should also satisfy the condition of the maximum value of the two
kinds of classification intervals. To meet these two conditions can make the confidence
and experience risks to achieve the minimum, so that we can guarantee the actual risk
of the system is the smallest
From low dimension space to high dimension space, the optimal classification line is
developed for the optimal classification hyper plane. The optimal hyper plane is subject
to the following constraints.
To meet yi=1 has the following conditions:
312 RISTI, N.º E5, 07/2016

Figure 2 – Maximal margin method.
(w ⋅ x ) + b ≥ 1 (7)
To meet yi=-1 has the following conditions:
( w ⋅ x ) + b ≤ −1 (8)
Comprehensive consideration of the above two factors will be obtained:
yi (( w ⋅ x ) + b) ≥ 1 (9)
The problem of maximizing the distance between the two planes is an optimization
problem:
1 1
min ϕ ( w ) = w = (w ⋅ w)
2
(10)
2 2
Lagrange multiplier is introduced to transform the problem into an unconstrained

optimization problem:
∑α ( y ( ( w ⋅ x ) + b ) − 1 )
m
1
L ( w, b, a ) =
2
w − i i (11)
2 i =1
Two extreme conditions:
∇b L ( w, b, a ) = 0 (12)
∇ a L ( w, b, a ) = 0 (13)
RISTI, N.º E5, 07/2016 313

Can get:
m
∑α y
i =1
i i =0 (14)
m
w= ∑α y x
i =1
i i i (15)
Calculated:
m m m
1
max a
2
∑∑a a y y ( x ⋅ x ) − ∑a
i =1 j =1
i j i j i j
j =1
j
(16)
m
s.t . ∑i =1
yi ai = 0
Figure 3 – The Schematic of SVM
3. Supply chain risk assessment index system

3.1. Internal and external environmental risks of supply chain
International economic environment risk refers to have a significant impact on the financial
order of the global economy, political structure, market environment and industry chain
structure changes quickly brought by the normal operation of the supply chain risk; risk of
the domestic economy is refers to the introduction of the relevant departments of the state
economic policy, industry regulations, and currency appreciation or craft value, inflation,
market fluctuation etc. of agile supply chain to produce adverse effects. The impact of
market volatility on agile supply chain is greater.
1. Market demand risk: With the rapid development of the market economy,
the buyer’s market has been formed in the global scope, the demand of
customers is gradually upgrading, showing a personalized, diversified and
advanced features. Reduce the update technology to accelerate the product life
314 RISTI, N.º E5, 07/2016

cycle becomes shorter and shorter, preservation of the product space is reduced,
which allows the management of awareness to the demand of the market, on the
market to forecast is more and more difficult, which in a large extent increased
agile supply chain operation risk.
2. Market supply risk: The normal operation of the agile supply chain cannot be
separated from the input of the external material; the supply market fluctuations will
have a serious impact on the agile supply chain system. Some of the supply chain
system must rising prices of raw materials, will lead to the rise in the cost of the supply
chain, in order to achieve a predetermined profit targets, we need to raise product
prices, which will allow customers to feel dissatisfied, may lose part of the market
3. Market competition risk: it refers to the agile supply chain facing increased
competition and brings the risk, such as competitors of the comprehensive strength
of promotion, the product innovation and the powerful competitors are likely to
break the supply chain management plan, the decline in the competitiveness,
losing part of the market, resulting in the agile supply chain loss.
Agile supply chain local risk refers to each decision within the body of a possible risk and
that suppliers, manufacturers, distributors and other supplies chain network all nodes
and links in operation process of supply chain overall unfavorable influence. The local
risk of agile supply chain includes human resource risk, technical risk and management
decision-making risk.
1. Human resource risk: One of the prominent features of the agile supply
chain is its agility, which can make a quick response to the market situation
and seize the opportunity to gain the competitive advantage in the fierce market
environment. It requires his employees with innovative thinking, need their all-
round development, high-quality staff team is an important guarantee for the
success of chain operation agile supply. Sensitive
2. Technology risk: Technology risk refers to the risk of the operation of agile
supply chain due to the progress of the society and the innovation of technology.
Agile supply chain is to reflect the market demand of, the new demand tend
to have a strong, consistent level of technology and equipment it is difficult to
produce customer requirements of products, so even if the time to seize the.
3.2. Design principles of supply chain risk assessment

Supply chain risk assessment is a complex process, it involves a lot of content and
factors, and each other are also related to each other, mutual tolerance. Therefore,
the construction of the evaluation index system to take full account of the relationship
between the influencing factors, eliminating the cumbersome factor, to prevent similar
factors. To establish a scientific and rational evaluation index system of agile supply
chain risk, not only can play a better role and function of evaluation, but also can make
the risk management level of agile supply chain greatly increased. In order to establish
a scientific and practical evaluation index system for the actual risk of supply chain
system, the following principles should be followed in the research process:
•• Scientific principle: The spirit and attitude of science is the basic attribute
of constructing evaluation index system. Only the index system which accords
RISTI, N.º E5, 07/2016 315

with the actual law can be established, which can reflect the objective reality and
the essence of things. In this text, to adhere to the scientific principle of building
the agile supply chain risk evaluation index system, and strive to reflect agile
supply chain in normal operation of main risk factors. In the evaluation process
obtained reliable, objective information. Finally, the appraisal results.
•• Overall principle: Because there are many factors that affect the normal
operation of agile supply chain, it is considered that the index number should
be considered when setting up the index system. Therefore, this article in
the choice of influencing factors, from all aspects of the impact of factors on
the evaluation of the impact of the object, and strive to comprehensive and
accurate list of factors, to ensure that each evaluation index is representative.
•• Measurable property principle: In the evaluation of the object, the
indicators should have the same evaluation criteria. The measurement index
system can be measured, but it is not that the index must be quantitative.
Quantitative indicators can be measured, and qualitative indicators can also be
measured by a certain method, means.
•• Relevance principle: Evaluation index system is a complex system, in the
process of building both to ensure that each index is independent of each other,
with a representative, but also to ensure that the correlation between indicators
is not denied. Each set of evaluation indicators have a certain internal contact, if
erase the contact, it will affect the evaluation results.
First order index Second order index Third order index
Macroeconomic form U11

Economic environment riskU1
External Market boom U12
environment risk
Policies and regulations riskU2 Affinity to relevant laws U21
Employee innovation capability U31

Human resource riskU3 Staff organization and communication skills
U32
Advanced technology U41
Internal Technical riskU4
environment risk Technical reliability U42
Market forecasting capability U51
Management decision riskU5 Production decision making ability U52
Organizational capability U53
Product reliability U61
Capacity control risk U6
Ability to deliver on time U62
Member Emergency delivery capability U71

Flexible risk U7
collaboration risk Order processing capacity U72
Information transfer distortion U81
Information risk U8
Information leakage U82
Table 1 – Index System of Supply Chain Risk Evaluation
316 RISTI, N.º E5, 07/2016

According to the characteristics and risk factors of supply chain, the risk assessment
index system of agile supply chain is established. The complexity of agile supply
chain, the index system involves many risk factors and some indicators is difficult to
accurately judge so as not to be able to index reliability quantitative analysis, but the
combination of the design principle of index system and from the agile supply chain’s
own characteristics, the evaluation index system has certain representativeness.
4.1. Data sources
When the training samples and test samples are constructed, it is necessary to emphasize
the typical and representative samples, which can fully represent the characteristics of
the agile supply chain risk in the study area and time. In view of this, we will study the
object is locked in the North China region has been 20 of the project manufacturing
enterprises as the core of the agile supply chain. Risk index score for: 0.1, 0.3, 0.5, 0.7,
and 0.9 five levels in the, followed by the corresponding risk level: low, in general, high
and higher. Agile supply chain the overall level of risk value set to between 0 and 1,
retained two decimal places, by agile supply chain risk reviews set as: [very low, low, in
general, high and higher], risk interval [0,1].
The process of obtaining the data through the expert investigation method is as follows:
1. Selected 30 supply chain management experts, including 15 research and
supply chain management of university professors and 15 enterprises engaged
in supply chain management in the middle of the above management.
2. Collection of data in the context of the agile supply chain and the operating
process of the agile supply chain.
3. Turns to seek advice. Send each expert a risk rating sheet and the relevant data
of the agile supply chain. Risk rating list requires experts to fill in the contents of
the various risk indicators, the value of the overall risk assessment of the supply
chain, as well as the evaluation of the value of the evaluation of the experts.
After the first round, asked the experts within the specified time to complete this
survey to recover, and organize the expert opinion; the second round, finishing
a good opinion questionnaire was sent to the experts, let the understanding of
the views of other experts, re of agile supply chain evaluation index and overall
risk score recover questionnaire and expert opinion; the third round, the second
round, finishing a good questionnaire feedback to the expert, based on the
questionnaire, the expert again scoring.
4. The questionnaire for the third round of the experts to deal with. Each risk
index of expert evaluation value, take all the risk values of agile supply chain
evaluation arithmetic average value.
4.2. Comprehensive evaluation

When the support vector machine is used to model the nonlinear regression, the selection
of kernel function has a great influence on the regression analysis. In this paper, we
choose the Gauss radial basis kernel function, also called the RBF kernel function:
RISTI, N.º E5, 07/2016 317

 x −x 2

 
(
K xi , x j ) = exp −
i
2σ
j
 (17)
 
 
Among them, σ is the kernel parameter, which represents the mean square deviation of
the Gauss kernel, that is, the width of the function in the direction of the independent
variable, the bigger the σ value, the wider the width of Gauss’s function will be. When the
width coefficient σ is small, radial basis function fitting performance better, but worked
as a small will cause the generalization ability is poor, so the kernel parameter σ is a very
important parameter. The process of optimizing the parameters of simulated annealing
process is shown in Figure 4.
Figure 4 – Simulated annealing process for parameters optimization experiment

Figure 4 illustrates the simulated annealing process for process parameter optimization
of support vector machine, because of the need to set some parameters of the simulated
annealing process, including initial temperature, fire rate, random disturbance radius to
generate new solutions, and the maximum number of iterations, this experiment used
the self-programming simulated annealing, which set the initial temperature for the
next 1, the ratio of temperature and temperature for the last 0.95, randomly generated
new disturbance radius is 0.2, set inside and outside circulation times were 20 and
100, the support vector machine parameter optimization is mainly to the optimization
process of the a penalty parameter C and kernel function, randomly generated in the
search range two values C and G, introduced into the training model of support vector
machine, support vector machine for training the fitness value, the fitness value of that
size of training error variance.
The last 5 samples were used as the test sample input system. The overall mean square
error of the model is obtained by testing: MSE=1.13225e-5, the smaller MSE value shows
that the model has higher prediction accuracy. The error of the predicted value and the
true value of the system is shown in table 2.
318 RISTI, N.º E5, 07/2016

Test sample 16 17 18 19 20
Expert evaluation value 0.42 0.44 0.37 0.42 0.28

SA-SVM output value 0.4087 0.4106 0.3934 0.4089 0.3008
Error 2.7% 6.7% 6.3% 2.6% 6.9%
Table 2 – Test results and relative errors of SA-SVM

The relative error of each test sample is very small, which indicates that the evaluation
model has good generalization performance. It can be seen from table 3 that the
evaluation results of the output of the simulated annealing support vector machine
model are in agreement with the expert evaluation results. Therefore, the model can
be applied to the agile supply chain risk assessment, and provides the decision basis
for the agile supply chain risk management.
Expert evaluation
Test sample SA-SVM output value Risk level Risk level
value
16 0.4087 higher 0.42 Low
17 0.4106 General 0.44 General
18 0.3934 General 0.37 Low
19 0.4089 General 0.42 General
20 0.3008 Low 0.28 Low
Table 3 – Evaluation results of SA-SVM and experts
4.3. Result analysis

In the risk factors affecting the normal operation of the agile supply chain, each factor has
different effects on the size and the probability of the risk of the system. The sensitivity of
each risk factor is analyzed by using the simulated annealing and support vector machine
model, and the key risk factors that affect the operation of the supply chain can be found.
Specific approach is in the same proportion from both positive and negative direction
change index for each risk factor value, then the change after the index value of input into
the model for risk prediction. The prediction results and original evaluation values are
compared, value at risk, more changes of risk factors more sensitive. All the risk evaluation
index values in positive and negative direction were changes in 15%. The changes of data
input into the trained simulated annealing - support vector machine model for a new
level of risk value, and compared with the original evaluation value, large differences in
the risk factors is the key risk factors. The results of the analysis are shown in table 4.
As seen from table 4 can supply chain risk evaluation index system of each risk factor
are the normal operation of agile supply chain produce certain effect, but each risk factor
of risk have varying degrees of role, to find out the key risk factors is agile supply chain
risk control is an important problem. From the point of view of the average deviation
table, risk factors: macro-economic form (U11), employee creativity (U31), advanced
technology (U41), reliability (u42), and market forecast ability (U51), on-time delivery
capability (U62), information transmission distortion (U81). Seven risk factors of agile
RISTI, N.º E5, 07/2016 319

New New
Index Risk Variation Difference mean
index index risk deviation
value value range (%) degree value
value value
+15 0.575 0.2873 0.0573 24.9

U11 0.5 0.23 21.1
-15 0.425 0.1903 0.0397 17.3
+15 0.575 0.2497 0.0197 8.5

U12 0.5 0.23 10.8
-15 0.425 0.1998 0.0302 13.1
+15 0.345 0.2742 0.0442 19.2

U21 0.3 0.23 20.2
-15 0.255 0.1812 0.0488 21.2
+15 0.345 0.3129 0.0829 36.0

U31 0.3 0.23 34.7
-15 0.255 0.1534 0.0336 33.3
+15 0.345 0.2491 0.0191 8.3

U32 0.3 0.23 11.0
-15 0.355 0.1988 0.0312 13.6
+15 0.575 0.3292 0.0992 43.1

U41 0.5 0.23 39.3
-15 0.425 0.1486 0.0814 35.4
+15 0.805 0.3184 0.0884 38.4

U42 0.7 0.23 39.7
-15 0.595 0.1359 0.0941 40.0
+15 0.345 0.3212 0.0912 40.9

U51 0.3 0.23 34.7
-15 0.255 0.1623 0.0677 29.4
+15 0.345 0.2533 0.0233 10.1

U52 0.3 0.23 11.8
-15 0.255 0.1992 0.0308 13.4
+15 0.575 0.2671 0.0371 16.1 16.4

U53 0.5 0.23
-15 0.425 0.1916 0.0384 16.7
+15 0.115 0.2572 0.0272 11.8

U61 0.1 0.23 10.1
-15 0.085 0.2109 0.0191 8.3
+15 0.805 0.3375 0.0925 40.2 35.7

U62 0.7 0.23
-15 0.595 0.1584 0.0716 31.1
+15 0.805 0.2631 0.0331 14.4

U71 0.1 0.23 14.2
-15 0.595 0.1978 0.0322 14.0
+15 0.115 0.2687 0.0387 16.8

U72 0.7 0.23 17.6
-15 0.085 0.1876 0.0424 18.4
+15 0.805 0.1618 0.0682 30.0

U81 0.5 0.23 28.0
-15 0.595 0.2968 0.0668 29.0
+15 0.575 0.2657 0.0357 15.5

U82 0.3 0.23 18.8
-15 0.425 0.1793 0.0507 22.0
Table 4 – Risk factors sensitivity analysis form
320 RISTI, N.º E5, 07/2016

supply chain operation to produce the effect of maximum, is the high risk factors. In of
agile supply chain management and risk control, to seize the key risk factors, focusing
on monitoring and management. It can not only greatly reduce the agile supply chain
management cost, but also can reduce the agile supply chain operation risk, increase the
agile supply chain profit rate.
5. Conclusions
With the increasingly fierce market competition, enterprises are facing the survival
environment of increasingly harsh, the client is also showing a diversified demand
for products, which makes the agile supply chain increases gradually in its operating
environment in the face of uncertainty factors, increasing the risk of, so agile supply
chain risk assessment research has very important practical significance. The of dexterity
affects the normal operation of the supply chain risk factors are analyzed, establishment
of agile supply chain risk evaluation index system, using simulated annealing - support
vector machine model of the risk evaluation, it indicates that the model in risk assessment
is applicable through empirical research. Combined with the agile supply chain has its
own characteristics, from agile supply chain of external environment risk, local risk and
members of the cooperative venture made a more detailed analysis, combined with the
basic principles of constructing the index system established agile supply chain risk
evaluation index system. Simulated annealing algorithm and support vector machine
(SVM) algorithm to do the detailed, in-depth analysis of the simulated annealing - support
vector machine evaluation model construction method, and illustrates the method for
agile supply chain risk assessment of feasibility and modeling steps. According to the
data obtained, the simulated annealing - support vector machine evaluation model were
trained and tested; the explanation on the choice of kernel function of the process of
construction of the evaluation model, the parameters of the model to determine some key
problems. The result of case analysis shows that the risk evaluation model established
in this paper is practical.
References
Informação, (16), 108–128.
Jeung, E., Daniel, Y. (2001). A conceptual model for interactivebuy-supplier Management,
relationship in electronic commerce. International Journal of Information, 21,
49–68.
Ming, J., Chuang, C. (2004). On a Replenishment Coordination Model in an Integrated
Supply Chain with One Vendor and Multiple Buyers. European Journal of
Operational Research, 159(2), 406–419.
RISTI, N.º E5, 07/2016 321

Prater, E., Biehl, M. (2001). International supply chain agility, tradeoffs between
flexibility and uncertainty. International Journal of Operations and Production
Management, 21(6), 823–839.
Suman, M., Patrick, T. H. (2004). Coordinating Supply Chains with Competition:
Capacity Allocation in Semiconductor Manufacturing. European Journal of
Operational Research, 159(2), 330–347.
Tah, J., Carr, A. (2000). Proposal for Construction Project Risk Assessment Using Fuzzy
Logic. Construction Management and Economics, (18), 491–500.
Vipul, A., Xiuli, C.(2004). Dynamic Balancing of Inventory in Supply Chain. European
Journal of Operational Research, 159(2), 296–317.
Zhimin, H., Susan, X. (2001). Co-op advertising models in manufacturer-retailer supply
chains: A game theory approach. European Journal of Operational Research, 135,
527–544.
322 RISTI, N.º E5, 07/2016

An Improved K-means Algorithm Application

in Distributed Data Mining and Information
Technology
Jianhu Gong*
*gongtiger@21cn.com
Computer Science and Engineering Department, Guangdong Peizheng College, Guangzhou 510830, China
Pages: 323–333
Abstract: With the development of computer technology in various fields, mass

data of various formats are generated and stored in the database. Mining useful
knowledge from massive data and applying it to the production practice has
become a key task, and distributed data mining provides a solution to this problem.
In this paper, the author proposes an improved K-means algorithm, and tests the
algorithm efficiency. The experimental results show that the improved K-means++
algorithm has significantly improved in clustering accuracy. Error rate reduce
to 10.7%, while K-means algorithm error rate as 17.3%. The improved algorithm
has good clustering quality, the running time is slightly lower than traditional
K-means++ algorithm, its advantage is cluster computing
Keywords: Improved K-means algorithm, MapReduce, distributed data mining,
clustering accuracy
1. Introduction
With the development of computer technology in various fields, a variety of formats
(pictures, music, video and documents, etc.) of the massive data generated and stored
in the database. Mining useful knowledge from massive data, and applying it to the
production practice of human society, has become a key task in the field of information
industry (Setiono, 1997; Liu, 2014). But too much data also brings a problem as data is
rich, but lack of effective information, it means the cost of knowledge is too high. Change
in demand driven technology. Distributed data mining provides a solution to this
problem. One of the key steps to realize the distributed data mining system is to transfer
the traditional data mining algorithms to the distributed computing platform (Lei,2008;
Jia, 2014). After decades of development, a variety of algorithms have emerged in the
field of distributed data mining, which have different needs and different tasks. These
algorithms can be improved to adapt to the new distributed environment, which is an
important step in the whole distributed data mining. In the real environment, the vast
majority of the data is not concentrated in a single server. They come from different sites
and are stored in a distributed database. And because of the limited computing power
of a single machine, the general data mining method is not very good for data mining
RISTI, N.º E5, 07/2016 323

An Improved K-means Algorithm Application in Distributed Data Mining and Information Technology
tasks (Jin, 2005; Freixo, 2014). Therefore, distributed data mining (DDM) technology
arises at the historic moment. As a mature technology, DDM has been widely used in a
variety of scenarios; it is a hot technology in the field of data mining. Distributed data
mining is used to solve the problem of data mining in homogeneous or heterogeneous
data sources. And parallel processing technology, which makes it very good to deal with
the scalability problem in the process of mining. It has been shown to have very good
efficiency in the distributed application environment.
Compared to centralized data mining, distributed data mining has a very good privacy
protection. In practical applications, due to privacy or legal considerations, a lot of data
is not suitable for centralized analysis. The use of distributed data mining, you can use
the computing resources of each node. Nodes will process the data and transmit the
intermediate results, not all of the data, so that it is natural to protect the privacy of the
data. In addition, the distributed data mining system will be deployed in large scale server
cluster, using software instead of hardware to ensure the stability of the system, thus
reducing the cost of the entire system. Clustering is based on the characteristics of the
data set; the data are divided into different categories (Zhang, 2008; Abreu, 2015). Its goal
is to make the data within the same class have a high degree of similarity, and the different
classes have nothing to do, and clustering is an unsupervised learning method. Among
them, the K-means algorithm is a classical clustering algorithm because of its simple
and efficient characteristics of a wide range of attention and research. In recent years, a
large number of research results have emerged on the K-means algorithm, including the
improvement of the efficiency of the algorithm and the optimization of the parallel system.
2. Improved K-means algorithm and its parallel implementation

2.1. K-means++ algorithm
Compared with the classical algorithm, the K-means++ algorithm is different from the
initialization strategy. Algorithm from the data set of random selection of the first center,
and the rest of the central point is through the probability to filter; the probability is
proportional to the distance between the node and the initial center. The pseudo code of
the algorithm is as follows:
Algorithm 1: K-means++
Input: k, X={x1,x2,x3…xn}
Output: C={c1,c2,c3…cn}
Step1: C←0
Step2: Randomly selected from the collection node x as the first initial nodeC=C∪{x}
Step3: while (|C|<k) do
Step4: Traversal x∈X, select the node as a new initialization center. C=C∪{x}
Step5: end while, |C|=k.
Step6: The element in the set C is used as the initialization center node, Lloyd’s iteration
according to the standard K-means algorithm, until convergence of the algorithm
The algorithm makes use of such an intuitive fact: a good clustering algorithm, the
cluster should be relatively dispersed. So in choosing a new cluster center should be
as far as possible to choose from the other center of the node. Both in theory and in
engineering practice, K-means++ has been proved to be far superior to the classic
324 RISTI, N.º E5, 07/2016

Figure 1 – Algorithm solving process
K-means. Although K-means++ algorithm has achieved great efficiency, but it also has
shortcomings, limiting its application, that is, the inherent nature of the algorithm.
When the input data set for the D dimension, the size as n, the number of clusters for
K, the total running time of the algorithm is O (nkD), which is similar to the time spent
on a single s’ Lloyd iteration. In spite of this, the algorithm can not be implemented
in parallel. In the initialization, a node can is chosen to be the probability of the i-th
center depends entirely on before I1 a center of the selection. The implementation of
the original K-means++ algorithm, can only run on a single machine, scanning the local
all the data to generate the initial center. When dealing with massive data, this limit is
even more serious. First, with the growth of the data set, the number of categories will
increase. For example, in a typical case, the algorithm has to scan all the data at once
to complete the initialization action by integrating the millions of nodes into 100 or
1000 clusters. If the residual algorithm steps (such as Lloyd ‘s iterative process) can
through parallel implementation, such as MapReduce to improve running speed, then
initialize spent time will become performance bottlenecks in the clustering process. So
in a lot of applications, the researchers used similar K-means++ other initialization
algorithm, they can achieve efficient parallel. In addition, in the K-meas++ algorithm,
the Euclidean distance is still used to calculate the distance of each data node to the
center point. In the calculation process, the different attributes of the node are treated
specially. However, in practical applications, some of the attributes must have a greater
impact on the clustering results, and some of the properties are weak.
2.2. Improved algorithm

Attribute weight is the influence degree of the clustering result. If the weight is large,
the influence of the attribute to the clustering result is bigger. This is more in line with
the practical application of the situation. In order to calculate the attribute weights, the
concept of information entropy is introduced. The information entropy to represent
objects contained in the amount of information, entropy is small, is included in the
greater the amount of information, distribution to the weights of the attributes should
RISTI, N.º E5, 07/2016 325

Figure 2 – K-means initial point selection
Figure 3 – Initial point selection of k-means++ algorithm
also bigger. Conversely, the smaller the amount of information, the smaller the value
should be. An input data set contains n number as m dimensional objects, the input data
can be represented as follows.
x11 x12 x1 m
x21 x22 x2 m (1)
X=

xn 1 xn 1 xnm
In order to calculate the attribute entropy and need with different dimension attribute
values can be compared with the various properties of 0-1 standardization:
n
bij = xij ∑x
i =1
ij
(2)
We can get the attribute weight matrix B
326 RISTI, N.º E5, 07/2016

b11 b12 b1 m
b21 b22 b2 m (3)
B=

bn 1 bn 1 bnm
The information entropy of the calculation of the j dimension attribute
n
sj = −p ∑b
i =1
ij ln bij (4)
Calculate the difference value of the j dimension attribute
cj = 1 − sj (5)
The difference value indicates the influence degree of the attribute to the clustering
result, when the difference value is Cj is smaller, then the Sj is bigger, and the influence
degree of the attribute to the clustering result is smaller. Calculate the weight of the j
dimension attribute:
cj
wj = m
(6)
∑c j =1
j
The weighted Euclidean distance formula of the object is expressed as:
∑w (x )
2
d ( xi , xk ) = j ij − xkj (7)
j =1
By multiplying the weights in the formula, we can describe the contribution of the
attributes to the distance between the objects. The greater the attribute value, the greater
the contribution, which is more in line with the actual application situation.
3. Hadoop and MapReduce programming framework

3.1. MapReduce programming model
Hadoop originated in the open source network search engine Nutch, and now has
become the Apache’s top open source projects. In 2003, Google published two papers on
distributed file system (GFS and MapReduce programming model, elaborated the inside
Google’s most important distributed storage platform GFS and distributed computing
framework, design idea. But only the main idea of the paper is not realized. In early 2006,
Hadoop Apache project was formally established, providing open source implementation.
It is based on the Java language development, designed to provide the programmer with
RISTI, N.º E5, 07/2016 327

an open source, universal distributed computing platform. Hadoop mainly consists of two
core components: a distributed file system HDFS and MapReduce programming model.
MapReduce model is one of the core components of Hadoop, and it is a distributed
computing framework, which is proposed by Google’s technical staff. In a scenario
where the problem is composed of mutually independent sub problems, it can maximize
the efficiency of the. MapReduce is divided into Map stage and Reduce stage, it hides the
complexity of the underlying system implementation, in a simple application, and users
only need to achieve two classes Mapper and Reducer can be achieved.
step describe
iteration Read the input file, traverse the data

map According to the function map () to deal with the input (key, value)
grouping Group according to the key value of the intermediate data.
reduce Will be the first step to get the packet protocol
iteration The final key to save on distributed file system
Table 1 – MapReduce process

The structure of MapReduce can be divided into four layers:
1. Runtime MapReduce, which is the basic part of the whole framework.
2. API Java layer, which provides the most basic programming interface, including
InputFormate, Mapper, Partitioner, Reducer and OutputFormat five classes.
They correspond to the five steps mentioned above. Hadoop to achieve some of
the commonly used InputFormat, Partitioner and OutputFormat, to facilitate
the rapid development of applications.
3. Tool layer, which is built on the API Java layer, including four tools: JobControl,
ChainMapper/ChainReducer, Streaming Hadoop, and Pipes Hadoop. Where
the role of JobControl is to deal with the DAG task. Streaming Hadoop provides
non Java language programming capabilities. Users can use Python or PHP and
other languages to write Mapper/Reducer. Pipes Hadoop is provided by the
C++version of the API. And the role of ChainMapper/ ChainReducer is to deal
with the chain task. Depending on the four tools, the tool layer provides users
with the ability to write MapRuduce operations and write complex tasks in a
non Java language. This improves the ease of use of Hadoop.
4. Application layer, located at the top of the system structure, to provide users
with the application.
Read the input file based on the InputFormat class provided by the user or by default,
and the input data are divided into slices. The number of slices is determined by the size
of the input file and the parameters given by the user. While slicing the logical structure
of the file, and each split corresponds to a Map task. File system will be copied to each
node, by the corresponding Map task processing.
328 RISTI, N.º E5, 07/2016

Figure 4 – MapReduce execution process
3.2. HDFS file system

Apache researchers have implemented the open source distributed file system HDFS
(Distributed File System Hadoop), and it as the default file system Hadoop. HDFS
and MapReduce closely together, constitute the basis of Hadoop. HDFS has the
characteristics of high fault tolerance and high expansion, which can be deployed
on a common commercial machine to provide reliable distributed file access for the
application of large scale data sets. It uses the classic Master/Slave structure, mainly
by the NameNode, SecondaryNameNode, DataNode and Client several parts. In most
cases, the cluster has a NameNode and a number of DataNode. The schema of HDFS is
shown in Figure 5.
Figure 5 – HDFS architecture
RISTI, N.º E5, 07/2016 329

Distributed environment compared with single environment more complicated and

HDFS has taken some measures, to improve the robustness, ensure in namenode,
datanode or network connection error, data storage can also return to the correct state.
1. fsImage and editlog damage: These two files are the system’s metadata mirroring
files and the file operation log, which belong to the core data of HDFS, so their
damage is very deadly. HDFS provides a number of copies, whenever fsImage
to Namenode and editlog for any changes, will be synchronized to each copy of
these data changes. While a number of copies of the operation will reduce the
performance of Namenode, but relative to the security of the core data, these
costs will be worth. And large scale data sets do not represent metadata as a
large scale. Whenever Namenode starts, it reads the most complete data from
the copy.
2. Load balancing: HDFS runs in a distributed environment, manages a large
cluster of machines, so it supports load balancing of data storage. If a file’s
access is greater than the set value, HDFS can create a new copy of the file to
reduce the load of the Datanode. If a Datanode free space is too low, HDFS will
automatically balance the load of multiple Datanode, and move the redundant
data to the idle machine.
3. Data integrity: Due to various reasons, such as equipment failure, network
connection failure, data stored in the Datanode may be damaged. HDFS check
and verify the means to ensure the integrity of the data. Whenever a new file
is generated or entered, the system will compute the verification of the file and
store the result in the same namespace as the file. Verify that the file has hidden
properties. When a data file is read, the system checks and documents under
the namespace to verify that it matches the data file. If the test does not match,
indicating that the data is corrupted, the system will read a copy from the other
Datanode (HDFS for each file to retain multiple copies, the default is 3, stored
in a different Datanode).
4. Heartbeat detection: Namenode through the heartbeat signal to detect the
operation of each Datanode. If you do not receive a signal sent by the Datanode
for a period of time, it is determined that the machine downtime, and take
corresponding measures. Data stored in the nodes labeled downtime of failure,
not to send them to a new read write instruction, so that some file block number
of copies will reduce the namenode. So Namenode takes a copy operation,
increasing the number of copies.
4. Experimental results and analysis

4.1. Experimental environment
In this experiment, 7 PC machines are used to form a distributed environment, each
machine’s hardware configuration is as follows: CPU as i3-4130 dual core, 2.6Ghz;
4GB memory; 100M Ethernet card; 500GB hard disk. The software environment is
as follows: Hadoop version: 2.5.0; JDK:OpenJDK 7; Mahout:0.9; operating system:
Ubuntu 12.04 64bit. Hadoop cluster for the Master/Slave structure, so that 1 PC
for the Master node, running NameNode, the other 6 machines for the Slave node,
330 RISTI, N.º E5, 07/2016

running DateNode, the implementation of the actual operation. Cluster partition as

shown in table 2.
Machine name User name LAN IP
Master CILAB 192.168.1.50

Slave1 CILAB1 192.168.1.51
Slave2 CILAB2 192.168.1.52
Slave3 CILAB3 192.168.1.53
Slave4 CILAB4 192.168.1.54
Slave5 CILAB5 192.168.1.55
Slave6 CILAB6 192.168.1.56
Table 2 – Cluster partition
4.2. Experiment of improved K-means++ algorithm

In addition to comparing the running time to measure the efficiency of the algorithm,
but also use the acceleration ratio, the expansion rate and other indicators to measure
the ability of the parallel algorithm. Acceleration ratio (Speedup) is an important
index to measure the parallel capability of the algorithm, which is the ratio of the
same computing task to the time consumed in a single machine environment and in a
distributed environment. The calculation formula is as follows:
Ts (8)
Sp =
Td
As an important measure of the algorithm for large scale data processing capability, the
extension rate (Scaleup) refers to the algorithm to deal with the operation time ratio of
small scale data and distributed environment. The calculation formula is as follows:
T (9)
Sc =
Tm * m
The validation of the algorithm is divided into two parts, the effectiveness of the algorithm
is that the quality of clustering and cluster environment to verify the performance of the
algorithm in the environment. In this paper, we use the Iris data set and the artificial
data set. Iris data sets from the UCI machine learning library, it is the pattern recognition
in the field of the most famous one data set. The use of randomly generated data as
an artificial data set to verify the parallelism of the algorithm. In the experiment, the
number of nodes is 5 artificial data sets, which are 2000004000008000001600000
and 3200000, respectively, and the dimension is 8 dimension. The data set in the
artificial data is consistent with the Gauss distribution. Using the traditional K-means++,
K-means, and the algorithm of this paper. The experiment was repeated 10 times,
RISTI, N.º E5, 07/2016 331

remove the minimum and maximum value, and the average value of the middle 8 times.
The experimental results are shown in table 3.
Clustering accuracy
algorithm
Average number Error rate average error
K-means 26 17.3% 0.2513

K-means++ 17 11.3% 0.1785
Improved algorithm 16 10.7% 0.1164
Table 3 – Comparison of clustering accuracy of various algorithms

According to the experimental results, we can see that both the K-means++ and the
algorithm of this paper have made a good improvement to the traditional K-means.
And in a number of repeated trials, the results of the traditional K-meas algorithm is
very unstable, relatively speaking, k-means++ and the results of this algorithm is more
concentrated. This is because the traditional K-means algorithm uses the randomization
of the initial center’s sake. Compared to K-means++, the proposed algorithm improves
the error rate by 7%, while the average error is reduced by 0.06. Because the method of
weighted distance calculation is used in this paper, the important attribute has a greater
impact on the clustering, which makes the clustering more accurate. Experiments
show that the improved algorithm has good clustering quality. The running time of the
algorithm is slightly lower than K-means++, it has the advantage of cluster computing.
5. Conclusions
After decades of development, a variety of algorithms have emerged in the field of
distributed data mining, which have different needs and different tasks. One of the keys
of distributed data mining system is to transfer the traditional data mining algorithms
to the distributed computing platform. According to the probability of the traditional
K-means++ algorithm to initialize the center point selection, compared with the random
selection of the method is more scientific, the efficiency of the algorithm is also compared
with the traditional algorithm has been greatly improved. But it is difficult to implement
the K-mean++ algorithm in the distributed system, and it ignores the difference of the
influence of different attributes on the clustering results. In this paper, the iterative
process of the algorithm is improved, so that it can be implemented on Hadoop, and
the implementation of pseudo code is also given, and the concept of attribute weight is
introduced to improve the quality of clustering. For the improved K-means++ algorithm,
the experiments show that it has good parallelism, and the clustering quality of the
algorithm is greatly improved compared with the traditional K-means algorithm. During
the experiment, it is found that the newly introduced parameter L has an influence on
the algorithm. After many times of comparative analysis, the effect of L=k/5 is the most
ideal. But at present, there is no theoretical basis for the selection of parameter L, which
is based on the experience of many experiments, which is the next research goal.
332 RISTI, N.º E5, 07/2016

References
Liu, B. (2014). Summary of distributed data mining. Journal of Hebei University of
Science and Technology, 35 (1), 80–90.
Jia, R., Yu,G. (2014). Parallel genetic K-means clustering algorithm based on MapReduce
model. Computer engineering and design, 2, 57–60.
Informação, (16), 108–128.
Jin, R., Yang, G. (2005). Shared Memory Parallelization of Data Mining Algorithms
Techniques. Programming Interface. IEEE Transactions on Knowledge and Data
Engineering,17(1), 71–89.
Setiono, R. (1997). A penalty-function approach for pruning feedforward neural
networks. Neural computation, 9(1), 185–204.
Lei, X., Xie, F. (2008). An emcient clustering algorlthm based on local optimality of
K-Means.Journal of Software, 19(7), 1683–1692.
Zhang, P., Wang, L. (2008). Method for initializing K-Means clustering algorithm based
on breadth first search.Computer Engjjneermg and Applications, 44(27), 159–161.
RISTI, N.º E5, 07/2016 333

Research on the Formation and Development

Mechanism of Brand Loyalty based on Social Media
Hongxin Li
*lch723@163.com
Logistics Trade College, Xi’An Eurasia University, Xi’An 710065, China

Pages: 334–346
Abstract: In recent years, with the popularization of Web 2.0 technology, smart
phones, tablet PCs and other internet terminal, China’s social media has made
rapid development, the number of social media users also showed explosive
growth. In marketing activities, pay attention to the deep emotional connection
with customers, create more value become an important source for enterprises to
improve customer loyalty and maintain sustainable competitive advantage. This
paper research the formation and development mechanism of brand loyalty based
on social media. Based on the empirical analysis, the result shows that social media
customer value has a significant positive impact on brand loyalty, and psychological
value (0.412) was slightly greater than functional value (0.365).So that the process
of customers’ communication with other person will increase the satisfaction and
dependence, then affect the brand loyalty.
Keywords: Social media, Web 2.0, brand loyalty, regression analysis, value
perception
1. Introduction
In recent years, with the popularization of the technology of Web2.0 development and
smart phones, tablet PCs and other mobile internet terminal, China’s social media made
the speed of development, the number of social media users also showed the explosive
growth(Butz, 1996; Sweeney, 2001). The popularity of social media not only has a great
impact on the enterprise brand marketing environment, but also changed the pattern of
consumer behavior, in accordance with the era of the AISAS consumer behavior model
(Holbrook, 2006; Freixo, 2014). In the background of this era, the consumer’s demand
and value perception also has a major change, only through the product, service and
other core values, more and more difficult to attract consumers. In marketing activities,
pay attention to the deep emotional connection with customers, create more value for
customers, and become an important source for enterprises to improve customer loyalty
and maintain sustainable competitive advantage. Based on this reality, more and more
enterprises realize the importance of social media and customer value.
Web2.0 is the development of the Internet to a new stage of the general term, represents a
new network model. In this network model, the majority of users with collective wisdom
and strength to break the centralized control of a small number of resource control,
334 RISTI, N.º E5, 07/2016

leading the entire Internet system. With the continuous progress of Web2.0 technology
and services, social media has been vigorous development, various types of social media
website, app, online tools emerge in an endless stream, attracting a large number of
users, in the field of commercial value quickly highlight(Wan, 2014;Abreu, 2015). In the
era of social media, a pair of many one-way communication mode change for many to
many network communication model, various types of social media application reveals
the attaches great importance to the user experience and take the initiative to create,
massive information integration classification, welcome the user’s active participation,
fully developed with users’ interest, open and inclusive, to stimulate the user maintain
a distinctive feature of the higher loyalty. Research institutions we are social to the
world’s key social, digital and mobile applications of statistics, and published research
reports(Wang, 2004; Jeffrey, 2006). The results showed that the total number of Internet
users worldwide has reached 30 billion, 72 billion of the world’s population in half using
mobile devices, nearly 20 million active social accounts, and last year compared to mobile
social media usage increased 23%. And all digital business will continue to grow at a
high speed. In China, the number of social media users also showed a sustained growth
of the state. 2016 China Internet Network Information Center (CNNIC) released the
China Internet development statistics report, the size of China’s Internet users reached
688 million, the annual total of 39.51 million people as new Internet users. Internet
penetration rate of 50.3%, compared with 2014 increased by 2.4 percentage points.
Figure 1 – Internet users change from 2005-2015 China

Along with the development of social media, various types of media platform emerge
in an endless stream, both to enhance the private network interaction, and in the
mutual sharing of the information resources, to the users create enormous value and
convenience, and have penetrated into the netizen of daily life. According to the statistics
of CNNIC Internet development in China, by 2015, 60.0% of Internet users on the
Internet sharing behavior holds a positive attitude toward, which very willing amounted
to 13.0%, compared to the reached 47.0%. In the information age, shift in user behavior
of network, enterprise’s marketing activity also occurred in the major changes. More
and more enterprises begin to use social media marketing to open up new customers
and improve the loyalty of old customers.
RISTI, N.º E5, 07/2016 335

Research on the Formation and Development Mechanism of Brand Loyalty based on Social Media
2. Social media
2.1. Social media concept
Mayrield Antony in his book according to the development of social media in the United
States at that time, the social media into the blog, forums, content communities, social
networks, blogs, six categories of wikipedia. Domestic Internet social media technology in
constant reference and imitation, has also been booming development. From Facebook to
all networks, from Youtube to Youku potatoes, from Twitter to micro-blog, from MSN to
QQ, various types of foreign social media form, in China has a corresponding localization
of the product. Scholars from different point of view, the social media members of the
huge be arranged. Through long-term observation and research consulting practice,
combined with China’s Internet Information Center (CNNIC) to disclose the authority of
data information, CIC released a survey of China’s social media pattern (Figure 2-1). The
report will be divided into the pattern of the pattern of “social marketing core platform”
and “consumer segments of the community of interest” two. Social marketing core
platform including instant messaging, music videos, blogs, microblogs, social network,
forum, mobile social networking, social life, e-commerce nine categories of; consumer
segmentation communities of interest, including: the 9 categories of social tourism,
dating, social business, corporate social, light blogging, social picture, Q&A short video
social, Wikipedia, social business platform.
Figure 2 – China social media landscape

Social media is different from the basic properties of traditional media, there is a general
consensus that the generated content user (UGC) and consumer generated media (CGM).
Based on social media user generated content attributes, social media platform, open, fair
face every Internet user, change the source of information, forming a huge user groups
involved in; based on social media audiences to form the properties of media , disruptive
changes in the mode of information dissemination and propagation direction, the social
media feature highlights out. These two attributes, coupled with the social media tools
336 RISTI, N.º E5, 07/2016

to see everywhere sharing, diffusion of elements, so that a wide variety of social media to
form a polymerization, connectivity platform. Therefore, this paper summarizes the core
characteristics of social media in the three aspects of specific elaboration.
•• Public participation: Due to the social media open given the right of every
Internet users to create and produce information, each individual can not only
as ever to receive the same information and could also become the creators and
the source of the dissemination of information. Ordinary users can freely express
their opinion, spread its recognition of the information without being blinded
by the strong tradition of media and one-sided, which challenges traditional
media information monopoly.
•• Community based communication: Social media has the characteristics
of two-way dialogue, the direction of information dissemination from the
linear structure into a network structure. Users at the same time of receiving
information also can give air to her comments on information, processing and
feedback; communicate with the main body of information dissemination.
Not only so, allow multi joint participation in the exchange of social media,
which have some common interests of the people to easily set up a community
or community, which in a certain extent to meet the people to expand their
interpersonal relationship, the needs of social contact.
•• Aggregate connectivity: aggregation is the combination of one or more content
together, and is one of the reasons why social media is so exciting, rapidly changing
and confusing. Today, most of the social media tools, sharing is one of the basic
functions. Through the attribute of the link, sharing, audience integration will a
variety of media fused together, enjoy collecting, producing, search, filter and a
dialogue with the non-barrier, and social media connectivity into full play.
2.2. Changes of social media to consumer behavior patterns

The meaning of AIDMA is in advertising and marketing, enterprises must consciously
do compelling, enable people to generate interest in the goods, and thus produce the
desire to buy, at the same time to let the goods a memory in the minds of consumers,
which eventually prompted this change in memory for the purchase of the line, in
order to effectively achieve the purpose of marketing. This model includes attention,
interest, desire, memory, and action. Reflects the traditional media environment of
the marketing relationship, the whole process of the purchase of the consumer can use
the traditional media, advertising, marketing, marketing tools. With the development
of the times, the popularity of social media, making the “information collection”
this link has been a huge change. Previously by the merchant control information
source and the release of asymmetric barriers to be broken, consumers are no
longer follow authority, no longer believe in advertising, and more willing to find the
views of the majority of people from a group of similar strangers. So the eyes and
ears of the consumers no longer be deceived, they have ability and the way to obtain
more realistic brand product information, “personal” and “experience” becomes more
important, more in line with the characteristics of the times AISAS model also arises
at the historic moment. AISAS mainly includes the attention, interest, search, action,
share 5 steps. At the attention stage, with the topic of detonation, event marketing,
viral and the way the product information to the information to the audience, cause
RISTI, N.º E5, 07/2016 337

the attention of potential consumers. Under the premise of attention, the potential
consumer’s interest will be inspired, began to take the initiative to use the network
search engine and other tools to collect more product related information. Through
further in-depth understanding of the contrast, make decisions, and even directly use
the convenience of the electronic business platform to complete the transaction action.
Finally, consumers can make their own shopping experience and product the most real
information released out, and more consumers to share information.
Figure 3 – Social media marketing AISAS model

Perceived value is consumer behavior of an important driving force, is customer decides
to purchase behavior and choice of the product key factors customer value is composed of
attribute, the destination layer and layer three levels, each level corresponding to a customer
satisfaction (as shown in Figure 4). Layer properties including product related to the specific
forms, characteristics, composition and other parts of the. Results layer is customers using
the product effect is expected. The target layer is the customer according to the use of the
effect of target customers the ability to form expectations. From the bottom to the point
of view, the customer will be according to the properties of products and service formed
for the expectations and the judgment result is to achieve their desired objectives; looking
down from the top, customers will according to their goals to determine products in the
use of scenarios the results of weight, adjacent to the three levels of hierarchy have mutual
influence. Enterprises can not only stay at the property level, but should pay more attention
to the result of the higher level and the satisfaction of customer value.
Figure 4 – Customer value hierarchy model
338 RISTI, N.º E5, 07/2016

3. Research model and hypothesis

3.1. Social media customer value and brand loyalty
In today’s era of social media, consumer behavior patterns have changed greatly. With
more serious homogenization of the goods, services, e-commerce platform provides the
distribution, sale and other supporting services more and more perfect, network payment,
privacy and security more and more secure, consumers in the AISAS mode of action stage
products, services and other core value has been relatively full attention and satisfaction.
Therefore, it is the key to obtain the advantage of the enterprise in the fierce competition of
the brand by establishing the deep dialogue relationship with the customers, to maximize
the additional value of the customer. First of all, the use of social media users can get the
heart of the entertainment to enjoy. Today’s social media tools, not only has pictures,
text, video and audio rich information content, also includes the many online test, small
game leisure projects, provide an easy way of entertainment to users by making full use of
the people of fragments of time. Second, the openness and participation of social media
give everyone a chance to show themselves fully. In the society of media customer value
research, in addition to the enterprise products and services to bring the core value, the
search for customers in the consumer behavior, share perceptual stage to add value to
become an important part of the customer value. These values ultimately affect consumer
brand loyalty. Based on this, this paper puts forward hypothesis as:
Hypothesis 1: the social media customer value is positively influencing the
enterprise brand loyalty.
H1a: social media customer value function value is positively influencing
enterprise brand loyalty.
H1b: social media customer value psychological value is positively influence enterprise
brand loyalty.
3.2. Social media customer value and brand trust

Trust is an important bridge to enterprises and consumers to establish long-term
relationship, the reliability of the brand is refers to the consumer that brand to the
ability to perform the promised value, is also the brand is to meet the needs of the
consumers. Brand intention is built on the basis of consumer beliefs, even if a certain
brand of products in the consumer process in some unexpected circumstances,
consumers are still able to support the brand. In this study, we are concerned in the
action stage before the user’s perceived value, that is, the user feel enterprises through
social media marketing behavior to them bring functional value, psychological value,
that is they feel “gains”. These gains will affect their attitude towards the brand, resulting
in satisfaction and trust. In the era of social media, brand loyalty is not only a repeat
purchase intention of a brand, but also the high degree of trust and recognition of the
brand’s products or services in the psychological and emotional. In this study, cognitive
and not based on consumer use of products, and consumers will not form evaluation of
brand attitude; but on the audience through social media platform and enterprise and
other audiences occurred. This attitude is the formation of the user brand trust and
the development of the brand loyalty of the psychological premise. Based on this, this
paper puts forward hypothesis as:
RISTI, N.º E5, 07/2016 339

Hypothesis 2: social media customer value is positively influencing enterprise

brand trust.
H2a: social media customer value function value is positively influencing enterprise
brand trust.
H2b: social media customer value psychological value of the positive impact of corporate
brand trust.
Summary of the scholars study, we found that brand trust and brand loyalty are related,
there is no trust loyalty, is fragile, brand trust is the foundation of brand loyalty. Based
on this, the paper puts forward the hypothesis 3:
Hypothesis 3: brand trust is positively impact on brand loyalty.
3.3. Scale design

Relationship between the study of social media customer value and brand loyalty,
antecedent variables is social media customer value, divided into specific functional value
and psychological value by one dimension. Because the variable is the customer’s brand
loyalty, and customer brand trust as an intermediate variable. In this paper, the research
results of domestic and foreign well-known scholars, combined with the research object
of this paper to fine tune the preparation of each scale indicators, to ensure that the
measurement of scientific and rigorous. The questionnaire survey content design using
the 5 component of the Li Ke special, so that the respondents to the various items on the
scoring. 1-5 points to quantitative index of each item, each number represents the mean
is the following: 1 represent “totally disagree”; 2 represents “disagree”; 3 represents “not
sure”; 4 represents the “consent”; 5 represents “totally agree”. 13 items to measure the
value of social media customer value and psychological value of the two dimensions, as
Item
Variable Survey item
number
QV1 Can obtain valuable brand, product related information
Function QV2 For brand, product information search is very convenient

value QV3 Have a certain impact on my purchase decision
QV4 Can help other users to understand the brand’s relevant information
QV5 To release pressure and relax.
QV6 Be able to pass the time when I am bored.
QV7 Can make me feel fun and happy
Psychological QV8 Free to express their ideas

value QV9 To satisfy my needs or opinions
QV10 Be able to feel a sense of self identity
QV11 Able to communicate with other people
QV12 Share the information I know, and get the support from others
Table 1 – Antecedent items
340 RISTI, N.º E5, 07/2016

4.1. Questionnaires
Using the questionnaire designed by the author to investigate and study, the final
effective questionnaire was 220, the following is a descriptive statistical analysis of the
test conducted by the effective questionnaire, by the statistical table 2,
Characteristic Effective Cumulative

type frequency
variable percentage percentage
male 119 54.1 54.1
Gender
female 101 45.9 100.0
Under 18 1 .5 .5
19-30 years old 141 64.1 64.5
Age
31-40 years old 69 31.4 95.9
Over 40 9 4.1 100.0
High school 8 5.9 5.9
Education Undergraduate 166 75.5 81.4
Graduate student 41 18.6 100.0
2000 below 26 11.8 11.8
2000 - 4000 60 28.6 40.9
monthly income
4000 - 6000 82 37.3 78.2
More than 5000 49 22.3 100.0
Table 2 – The questionnaire statistical analysis
4.2. Descriptive statistics

Descriptive statistics is the income survey data collecting, classifying and simplified
plotted as a graph or chart, through the observation and analysis of the data, describe and
summarize the relationship between the data and the features of the variables of a statistical
method. Descriptive statistical analysis includes the mean, frequency, standard deviation,
the main focus on the data of the trend, the degree of dispersion and the intensity of the
description and analysis. The mean is the average level of the sample in each variable;
the frequency is the distribution of the observed data; the standard deviation response is
the degree of dispersion of the data set. This study uses SPSS17.0 software to collect the
data of the independent variables, intermediary variables, the results of the survey results
of variables descriptive statistical analysis, the results of the specific table 3.
4.3. Reliability analysis and validity analysis

Reliability refers to the degree of consistency of the results obtained from repeated
measurements of the same object using the same method. To do reliability analysis is
mainly to determine the survey questions in the questionnaire to investigate the purpose
and intent of the degree of reaction, the survey data is true and reliable.
RISTI, N.º E5, 07/2016 341

variable projects mean value overall average

3.70
3.72
Function value 4 3.68
3.73
3.55
3.54
3.76
3.66
3.54
Psychological value 8 3.63
3.76
3.54
3.65
3.57
3.45
3.56
Brand trust 4 3.57
3.60
3.67
3.81
3.78
Brand loyalty 5 3.66 3.74
3.80
3.63
Table 3 – Summary statistics
Cronbach’s α Standard term Cronbach’s α items

.931 .932 22
Table 4 – The reliability statistics
variable Cronbach’s α Reliability level items
Function value .729 high 4

Psychological value .878 high 8
Brand trust .759 high 4
Brand loyalty .838 high 5
Global scale .931 high 21
Table 5 – Variable reliability coefficient
342 RISTI, N.º E5, 07/2016

In this paper, SPSS17.0 software is used to verify the validity of the data by factor analysis.
In the measurement process, the KMO and Bartlett are measured by measuring the
spherical degree of the questionnaire data to analyze the adaptability of factor model.
When the variable is suitable for factor analysis, the principal component factor is
extracted, and the variance of the maximum rotation method is used to select the factor.
KMO and Bartlett
Kaiser-Meyer-Olkin .923
Bartlett’s test Approximate chi square 1042.005

df 78
Sig. .000
Table 6 – Customer value validity test
4.4. Regression analysis

In determining the relationship between the variables and need to use regression
analysis to determine the quantitative relationship between variables and judgment
of different variables affecting the degree size, determine the interacting relationship
between variables. In the results of the SPSS output, you can report to the R side, F test
values and Beta values to explain the results. The value of R is between 0 and 1, which
indicates the degree of the explanation of the dependent variable in the equation, the
greater the value, the stronger the effect of the independent variable on the dependent
variable. F test, the linear relationship between regression equation and the regression
equation was significant. Generally speaking, the significance level is above 0.05.
Non standardized Standard

Model coefficient coefficient t Sig. F Sig.
B SE trial
constant 1.424 .204 6.980 .000 140.281 .000
1 Psychological
.651 .055 .641 11.844 .000
value
constant .709 .228 3.104 .002 96.844 .000
Psychological
.418 .066 .412 6.363 .000
2 value
Function
.414 .073 .365 5.645 .000
value
Table 7 – Customer value and brand loyalty coefficient

Two dimensions of the independent variables into the model, adjusted R value reached
0.487, social media customer value explains 48.7% of the brand loyalty degree of variation;
F and T values were significantly, confirmed the significant regression equation of linear
relations and coefficients, so socialization media customer value on brand loyalty and
positive effect significantly. And observation of the two dimensions of the independent
RISTI, N.º E5, 07/2016 343

variables of the standard coefficient Beta value, psychological value (0.412) is slightly
greater than the value of function (0.365), it can be inferred that the impact of customer
psychological value on brand loyalty is slightly greater than the value of the function.
According to the results of the data analysis, social media customer function value and
psychological value have positive impact on brand loyalty.
Intermediary effect refers to the variables and for the relationship between the variables
is not direct causal chain relationship but through the intermediary variables indirectly
influence the generation of and independent variables through the intermediary variable
indirect effect on the dependent variable said as the intermediary effect. If in the model
also considered independent and mediating variables due to the influence of variables
and with independent variables separate effects due to variable results in comparison,
we can come to the intermediary variables of the intermediary is completely mediating
role, or part of the intermediary role. The above analysis has shown that the social media
customer value of the independent variable and the dependent variable brand loyalty
and the middle variable brand trust have significant positive correlation. The results are
dependent independent Standard Standard

B t Sig.
variable variable error coefficient
Psychological
.418 .066 .412 6.363 .000
Brand loyalty value
Function value .414 .073 .365 5.645 .000
Brand trust .534 .060 .517 8.945 .000
Brand loyalty
Function value .362 .065 .319 5.522 .000
Brand trust .447 .061 .433 7.382 .000
Psychological
Brand loyalty .267 .062 .263 4.326 .000
value
Function value .230 .070 .203 3.302 .001
Table 8 – Brand trust intermediary role

According to the results of data display, and gradually the function value and brand trust
into the regression equation, regression coefficients are still significant, and beta value
decreased from 0.41 to 0.36, visible, and the independent variables due to significant
variables that influence weakened that brand trust in the value of the function on the
brand loyalty is related to play a partial intermediary role. The same, and gradually the
psychological value and brand trust into the regression equation, model coefficients are
still significant, and psychological value beta value also declined, indicating that brand
trust in the psychological value of brand loyalty is related also play a partial intermediary
role. Finally, the brand trust and social media customer value of each dimension into
the stepwise regression equation, functional value of the impact of the brand loyalty
significantly weakened, but still less than 0.05, the coefficient of the model is significant.
Therefore, we can see that brand trust plays a partial intermediary role in the positive
impact of social media customer value on brand loyalty.
344 RISTI, N.º E5, 07/2016

5. Conclusions
According to the results of empirical research, we can know, in today’s era of Internet
users, in the use of social media, whether active or passive, more or less will come
into contact with many enterprises brand marketing activities. Through social media,
customers can not only by understanding the convenient information search, links to
the rich information of brand products, and enterprise or other users of communication,
exchange, share, access to the functional value of the external and internal psychological
value, and thus to attitude of corporate brand recognition and trust, the creation of brand
loyalty. In the social media customer value on brand loyalty effect of two levels in, customer
value of the psychological impact on the brand loyalty of the slightly larger than the
value function, which may is because when the customer use social media to participate
in the enterprise marketing activities, and enterprises or other users to communicate,
share the process, easier access to social, self-value of psychological satisfaction, and
thus more easily to the enterprise produces satisfactory and dependence, thereby affect
brand loyalty. In social media customer value on brand trust, customer value of beta
coefficient slightly larger than the psychological value, showing the user brand trust has
the more important depends on to the customer on the brand, the real information of
products to fully grasp the and comparative analysis.
Brand trust is an important factor affecting brand loyalty. At the same time, the social
media customer value can not only influence the brand loyalty, but also influence the brand
loyalty through brand trust. Enterprise in the marketing practice, attach importance to
customer value, in fact, is to require companies to ensure good quality of own brand
products and services, protect the customer’s core values are to meet at the same time,
pay full attention to the additional value of the psychological level of modern customer,
customer value and brand trust complement each other together for enterprise creation
and cultivate brand loyalty. Anyway, Internet users in in the enterprise in the society of
media platform for brand marketing activities to get the function value of the external
and internal psychological value from, through the perception of these values, the user to
the enterprise brand cognition and attitude will change accordingly, produce for brand
affection and trust, the trust attitude will be in the final stage of the consumer behavior
model implementation into the brand loyalty, the enterprise marketing activities to
obtain the final real returns. Social media has revolutionized business communication.
Pay attention to the value of social media marketing, to maximize the value of the core
value and additional value of customers, in order to enable customers from the heart of
the recognition of corporate brands, brand trust, create brand loyalty.
Acknowledgments
The work of this paper is supported by Education Department of Shaanxi province
(15JK2176).
RISTI, N.º E5, 07/2016 345

References
Informação, (16), 108–128.
Butz, H., Goodstein, L.(1996). Measuring customer value: gaining the strategic
advantage. Organizational Dynamics, (3), 24–30.
Holbrook, M.B. (2006). Consumption experience, customer value, and subjective
Personal introspection: An illustrative Photographic essay. Journal of Business
Research, 59, 714–725.
Jeffrey, W., Overbya, E.L. (2006). The effects of utilitarian and hedonic online shopping
value on consumer preference and intentions. Journal of Business Research,
59(10), 1160–1166.
Sweeney, J.C., Soutar, G. N. (2001). Consumer Perceived value: The development of a
multiple item scale. Journal of Retailing, 77(2), 203–220.
Wan, G. (2014). Review and Prospect of research on consumer brand trust. Modern
management science. 8, 55–58.
Wang, Y., Fesenmaier, D. (2004). Modeling Participation in an Online Travel Community.
Journal of Travel Research, 2(42), 261–270.
346 RISTI, N.º E5, 07/2016

Research on Social Support System of Peasant

Workers’ Citizenship based on Data Mining Theory
Ji Liu, Xiaomei Zhang , Zhilin Suo *
*77291063@qq.com
Northeast Agricultural University, Harbin 150030, China

Pages: 347–359
Abstract: With the development of the Internet, the information society has
entered the era of big data. In this paper, the authors analyze the social support
system of peasant workers’ citizenship based on network and data platform.
Through the network investigation, the result shows that the building of social
support system plays an important role in promoting citizenization of peasant
worker. The main body social support can be resolved as two dimensions: internal
support system and external support system. The internal support system is
also divided into governmental support and non-governmental support. And
the external support system is based on the support of kinship and the support
of geopolitical relationships. In order to structure the social support network
system of peasant workers, government need stimulate the non-governmental
organization, kinship and geopolitical relationships. At the same time, we need
pay attention to the three links when peasant workers integrate into society, as
countryside exiting, city entering and city integrating.
Keywords: Data mining, social support, network data, peasant worker
1. Introduction
In the world, every country has a problem about urban integrating of immigrant. And
because Chinese present stage is in the transition from traditional agricultural society to
industrial society, peasant workers constitute the mainstream of the floating population.
In other countries, floating population entering into cities and settling down cities is
at the same time. The system of Urban and rural binary household registration which
formed in particular historical period makes peasants of having entered into cities unable
to enjoy the urban residents welfare. When peasant workers are in changing citizen,
there unavoidably emergence contradictions in economy integrating, culture integrating
and mentality integrating(Tao, 2015; Zhong, 2015). The problem of social integration in
immigrant has aroused the attention of scholars both at home and abroad, in areas such
as economics sociology, psychology, demography and sociology(Su, 2014; González,
2015). Chicago school at the end of the 19th century in the United States begin to pay
close attention to foreign peasants social integration, conducting research in the field of
sociology, there are a number of scholars, respectively, the concept of social integration
and social integration of standard, the countermeasures of social integration, variables
RISTI, N.º E5, 07/2016 347

Research on Social Support System of Peasant Workers’ Citizenship based on Data Mining Theory
affect the social integration and social integration after the question has carried on the
thorough discussion. After absorbing the western research, Chinese scholars has carried
on the analysis and interpretation the problem of peasant workers urban integration
in present stage(Xu, 2011; Qiu, 2011).CAI fang, according to the law of economic
development from the macro, came to the conclusion that labor transform from the
agricultural sector to industrial sector is inevitable process. Inside of peasant workers
appear heterogeneity and the group marginalized in the whole society (Shen, 2012).
Chen investigated the result of peasant workers integrating city, he thinks peasant
workers into the economy and social integration can promote the peasant workers
group participation(Cheng, 2013). Overall, these studies used qualitative analysis and
quantitative analysis method. From the content point of view, there have summarized
the process of integration of immigrant city problem and speculate about the future, for
peasant workers have segregation analysis, but there aren’t from internal motivation
and external stimulation on comprehensive analysis of rural peasant workers, in turn, to
promote the realization of the peasant workers’ citizenization of research. Therefore, in
this paper, using the theory of sociology and psychology social support system, providing
external support for peasant workers are analyzed the main body of government and
nongovernmental organizations, internal support system for geopolitical relations
and kinship organizations. we need notice three level, Macro, meso and micro level of
support. There are three links when peasant workers integrate into society need pay
attention to, countryside exiting, city entering and city integrating(Freixo, 2014).
Social support system is also called social relationship network, the concept is put
forward by the sociologist in the 1970s, it refers to a relatively stable system of network
which formed in people society life and was self-center to support access to the survival.
This system is a particular system which is often considered as providing material
and spiritual aspects of the voluntary help to the social vulnerable groups, to help
disadvantaged groups and maintain social stability plays an important role. Society
builds an atmosphere one or parts people are loved, needed and can realize the self-value,
the atmosphere is a kind of strength or factor which promote human developing in social
environment(Lu, 2012).Nowadays, in the field of sociology researches for low-income
families in urban areas, the elderly, women, college students and other groups, these
studies including the research on social support network of urban poor social support
network . There are also a cross-cultural comparative study with the foreign community,
such as the research on comparing the social support network of the elderly in Beijing
and Liverpool. The research object of these studies is our country special national
condition. These studies used the western sociological theories and methods to explain
our country problems at the present stage and enriched and developed the theory, for
the Chinese sociology research entering into the world and carrying out international
academic dialogue laid a solid foundation. Generally speaking, the social support refers
to getting help from others in society. And the research of social support network is an
important branch of social network analysis, the individual’s social support network is
a network that aims to get all kinds of resources to support, such as money, emotion
and friendship, it’s like researching social network that you can study social support
network from two aspects: the overall network and individual network. Since the 1970 s,
western scholars began to study social support networks. Recently, overseas research on
social support network mainly includes the composition of social support network and
348 RISTI, N.º E5, 07/2016

social support network how to provide social support to individuals. Then we make
analysis and comparing for social support network characteristics between different
national, regional, ethnic and different groups.
2. Data mining
In today’s society is an era of information explosion, how to make full use of all kinds
of information for human services has become increasingly important. As a burgeoning
knowledge discovery technology -- data mining and decision support tool for one decision
support system has attracted more and more attention, which people from large amounts
of data to obtain useful and interesting information provides a convenient way. Large
number of branches, classification and prediction has been a very attractive main branch
in data mining; many scholars from different angles such as statistics, information theory
and neural network do a lot of research, made a series of important results.

The history of the development of computer technology is also the history of
automatic data acquisition, generation and processing in a sense. At present, the
global service database for all walks of life at least a few million, the database is not
only the number of rapid growth, but also more and more large scale. For example,
a series of earth observation satellites launched by the United States Department
of space and earth observation satellite data is about 300 thousand GB. Faced with
such a huge data resources, the general management system of the query retrieval
mechanism and statistical analysis method has been unable to meet the needs of
the application. Data mining is an important research field that is produced and
developed rapidly under the background of the application requirement. Through
the analysis of the data, data mining can discover the important data model, and
make a great contribution to the business decision-making, knowledge base, science
and medical research.
One of the salient features of data mining is that the extracted patterns are implicit and
unknown. This is different from the expert system is only to have knowledge of learning.
And these new knowledge in an easily understood form, the mining process is a person
machine interaction and spiral process, it needs to repeatedly adjust, in order to dig
out the knowledge quality taller and more effective. K - means algorithm accept input
RISTI, N.º E5, 07/2016 349

Figure 2 – Knowledge discovery process
k; Then could be divided into k n data object clustering in order to make the clustering
obtained satisfy: clustering objects in the same similarity is higher; Unlike smaller
clustering objects in the similarity. Cluster similarity is the use of the cluster of objects
in the average obtained a “central object” (center of gravity) to calculate.
Figure 3 – K-means algorithm
3. Peasant workers’ social support network system

3.1. Decision tree algorithm
Chinese peasant workers in cities are suffering multifaceted social exclusion problems
about economic, political, public service and social relations. Facing with Problems of
social exclusion of peasant workers, governments at all levels should take comprehensive
and systematic countermeasures, to eliminate social exclusion of peasant workers
350 RISTI, N.º E5, 07/2016

progressively. In view of our country for thousands of years of farming culture, Combined
with the limit of binary system, peasant workers’ social relations formed a characteristics
of nature, involution, and localism. Although now reduces the threshold of farmers to enter
the small-towns, and fine-tune the household registration system in part of major cities,
and thus weakens the social impact of the household registration system, but its history
inertia effect for major cities’ managers still has very big effect. And this kind of influence
constraints to peasant workers in urban living space and activity space, makes peasant
workers disadvantaged groups, in the city but not into the city and in a tough position.
In order to study the effect of social support for rural peasant workers merging into
the urban communities, research group conducted a questionnaire survey based on
peasant workers’ basic situation, in four main district in Harbin, this survey involves
the Daoli,Daowai, Nangang and Xiangfang district of Harbin, and distribute to 1000
peasant workers, collecting 967 valid questionnaire, with 96.7% efficiency. The
questionnaire mainly involves peasant worker’s gender(X1), age(X2), occupation(X3),
marriage condition(X4), and education level(X5),etc. From these aspects to investigate
the urbanization of peasant workers, deals and analyzes with the questionnaires used
and sorts data in the questionnaire with SPSS22.0 SOFTWARE. The basic situation of
the survey is shown in Table 1 as follows.
Name Symbol Definition Symbol Number Percentage

male X11 632 65.4%
Gender X1
female X12 335 34.6%
Under-20 X21 246 25.4%
21-30-years X22 351 36.3%
Age X2 31-40-years X23 220 22.8%
41-50-years X24 97 10.10%
Over-50 X25 53 5.40%
Construction X31 323 33.40%
Service X32 195 20.17%
Occupation X3 Manufacturing X33 246 25.44%
Transportation X34 119 12.31%
The other X35 84 8.68%
Married X41 497 51.41%
Unmarried X42 393 40.64%
Marriage X4
Divorced X43 42 4.34%
Widowed X44 35 3.61%
Junior High X51 589 60.91%
Senior high X52 337 34.85%
Education X5
College degree X53 38 3.93%
Bachelor degree or above X54 3 0.31%
Table 1 – Survey on basic situation of peasant workers in Harbin City
RISTI, N.º E5, 07/2016 351

According to the preceding basic situation of the investigate and the descriptive statistical
analysis, conduct an empirical analyzing on the aspiration of urbanization of peasant
workers with binary logistic regression model. Binary logistic regression model were
an effective tool which uses to analyze the relationship between qualitative variable
and influencing factors, logistics regression model doesn’t require the explain variables
or explained variables obey the multivariate normal distribution, it is based on the
probability, for qualitative data it is usually use 0 to express nothing happens, and use 1 to
express the event happens. According to the survey, the questionnaire about the effect of
the aspiration of urbanization of peasant workers was designed with respect to gender, age,
marriage condition, education level, urban income and work environment, assume that
the willing of peasant worker number i is Pi, the urbanization influence factor of peasant
worker j is Xj. Build logistic model to analyze the 967 samples, the model’s general type is:
n n
Pi = F (α + ∑β x
j =1
j j ) = 1 / {1 + exp[−α + ∑β x
j =1
j j )]}
Pi = F ( Age, Gen, Mar ,Inc, Edu, Empi ) + e
Using SPSS20.0 software for data processing, the specific variables of the model and the
empirical results are shown in Table 2.
The variable symbol Estimated coefficients Sig value

gender Y1 -0.18 0.305
age Y2 -0.304 0.002
Nature of employment Y3 0.336 0.003
Marriage condition Y4 -0.227 0.001
Income situation Y5 0.315 0.003
Education level Y6 0.701 0.001
Table 2 – Logistic model estimation results will of people of peasant workers

According to the result, age(Y2), nature of employment(Y3), marriage condition(Y4),
income situation(Y5),education level(Y6)have a significant effect on the urbanization
aspiration of peasant workers, but there is no significant effect on gender variables.
There was a negative linear correlation between gender variables and the urbanization
aspiration of peasant workers, and the urbanization aspiration diminishing with age. This
is because the peasant workers at the low age seem more quickly to accept new things of
the city, more quickly into the city life, so the will of the urbanization is higher. And high
age of peasant workers have been used to the existing way of life, it is not easy to change.
So the age of high willingness to urbanization of the peasant workers is low. There was
a positive linear correlation between nature of the employment and urbanization of
peasant workers. Marriage condition variables and urbanization of peasant workers will
become negative correlation relationship. Compared with the married peasant workers,
unmarried, divorced, widowed peasant workers have more urbanization willingness.
352 RISTI, N.º E5, 07/2016

Because of the married peasant workers are greatly influenced by the family, constraints
are also relatively high, so the urbanization willingness is low. There was a positive linear
correlation between income levels and the urbanization willingness of peasant workers.
And high income can make peasant workers live like citizen people. Hoping to enjoy equal
respect and treatment with the citizens, and more willing to blend in and enjoy the city
life. Such a high income can increase the willingness of migrant workers in the city. There
was also a positive linear correlation between education condition and the urbanization
willingness of peasant workers. Peasant workers from the higher education level, cultural
quality is relatively high, the more easier to adapt to urban life in the city, the more
employment opportunities in the city,the more easier to produce in the city. Therefore,
the higher education level of migrant workers is higher.
Network size is measured with the number of individual’s social support. In the interview,
the network of the different types of support scale range, from 0 to 5 people, the overall
network size between 2-3 people. In other words, each person can at least get social
support what they need from 2 or more people.
Network size distribution(%)

Dimension
average size
type
0 person 1person 2person 3person 4person 5person
Practical
1.84 0 56.3 26.9 9.6 4.3 2.9
support
Emotional
1.51 4.2 71.8 18.6 3.2 2.2 0
support
Social support 1.13 10.4 56.1 12.9 7.8 7.4 5.4
Table 3 – The size and distribution of peasant workers’ social support network
What is clear from the data is that 83.2% new generation peasant workers can get
practical support from 1 to 2 people, 90.4% new generation peasant workers can
get emotional support from 1 to 2 people,69% new generation peasant workers can
get social support from 1 to 2 people. In particular,4.2% people don’t have someone
to provide emotional support,10.4% of people don’t have someone to provide social
support. According to the information in the process of the interview, suggest that the
people who don’t have anyone to provide emotional support are male, and the 6 people
who consider no one can provide for social support, five of them are female. To examine
minutely, nobody on earth is to provide support for yourself, or no one they want to put
forward to the support request, respondents believe should be the latter, no emotional
support men believe “Himself in the village are celebrities who jumped farming door,
so they are unwilling to complain to parents seeking comfort, can’t afford to lose face”,
and no social support women think “everyone’s economic capability is not the same, we
can’t go together “and” We are all classmates, but now everything is inferior to others “is
the cause of the lack of social support. The composition of the social supporting network
relation, mainly refers to the main constitution of the specific relationship between main
research objects and the network members, in this study the mainly set relationships
include kinship and unrelated relationships, which kinship consists of parents, siblings,
spouse or lover, other family members. An unrelated relative includes students, fellow
colleagues, friends and net friend. Table 4 the specific relationship.
RISTI, N.º E5, 07/2016 353

From the table 4, there are practical supporting network, emotional supporting network
and social supporting network as the compositions of the new-generation peasant
workers’ socially supporting network, the proportion of unrelated relationship is higher
than that of kinship .The above data shows that new-generation of peasant workers’
social supporting network is different from the first- born social supporting network with
blood and affinity as a core in the early stage of the flow of peasant workers, after-born
social supporting network is increasingly influencing new-generation peasant workers,
of which both in the condition of co-existing and complement each other. It can be seen
from the composition of actual supporting network relations that in all the relatives,
brothers and sisters shares 15.4%, spouses and lovers accounts for 14.5%, with parents
accounted for only 10.7% , which means that a new-generation of peasant workers are
less dependent on their parents in the actual supporting.
The dimension type and percentage

Relationship types
Practical support Emotional support Social support
All family numbers 43.2 44.1 23.1
Parents 10.7 2.8 0
Brothers and sisters 15.4 11.2 12.3
Spouse or lover 14.5 27.5 8.8
Other relatives 2.6 2.6 2
All the relatives 56.8 55.9 76.9
colleagues 8.4 5.2 3.5
classmates 20.6 17.3 30.2
Friends(net friends) 25.8 30.9 41.1
Person from the same
2 2.5 2.1
village
Table 4 – The size and distribution of peas

But during the interview, almost all respondents have admitted “ at least half a year’s
time after working they still ask their parents, brothers and sisters for money, or
you can’t live in the city “.At the same time, interviewees with children still depend
on their parents, because they could not bear the economic burden brought by no
working but looking after children at home. In related relationships, the proportion
of students is 20.6%, proportion of friends is 25.8%.Different from other studies,
higher proportion of student sexists in this study which is almost consistent with the
proportion of friends. As can be seen from the structure of the emotional supporting
network relations, in all the kinship, spouses and lovers shares 27.5%, brothers and
sisters accounts for 11.2%, only 2.8% for parents, from which it can be seen that a
percentage of emotional attachment to parents is very small. Respondents all believe
that “parents get older and older with not being able to know a lot of things and
you neither can make them understood nor get their truly useful suggestions, so it’s
unnecessary to let them worry about”.
Spouses and lovers in emotional support network accounts for a larger proportion
of kinship, which also explains the marital status of the respondents from one side.
354 RISTI, N.º E5, 07/2016

In this interview, married respondents or those with any engagement are engaged
or married in the first half of year after the graduation or just graduated, both of
them are graduating student or graduated students from the same village, of whom
wedding and engagement reasons are anyway that all want to work in urban areas in
the future, if they get married or engaged, there will be a person to rely on. What this
new related relationship brings to peasant workers distinguishes from the support
of parents, brothers and sisters. In the unrelated relationships, the proportion of
students is 17.3%, friends are 30.9%, where the net-friend has played a very important
role in, they think in the virtual world that “you don’t need to disguise and just speak
out their boredom without worries”. Summarizing table 3 and table 4, it can be seen
that the social support network of the new generation of migrant workers is small,
and some social support is not to be obtained but not to be obtained. In three kinds
of supporting networks, unrelated relationships occupies more proportion than
kinship, illuminating that for the new-generation of peasant workers, their mainly
social supporting is derived from the unrelated relationships. Practical support for
the new generation peasant workers from parents is greater than the emotional
and social support while social or emotional support from friends is greater than
practical support. Social supporting networks consist mainly of parents, siblings,
friends, classmates, which has the characteristics of homogeneity.
4. Provide social support for rural migrant workers

4.1. Design of experiential teaching method
The main social support providing migrant workers can be divided into external and
internal support; external support means that the support of the government and non-
government support, internal support are based on kinship support and geopolitical
relations support, which shows in the table.
supporting main body sub-categories concrete content

household registration reform
Promotion of
political status security of migrant workers to participate in the
community’s election campaign
Make sure migrant workers into the urban medical,
Increase economic unemployment, pension, work injury and other
income insurance coverage
government support Urge employers to timely and full payment of salary
provide free high quality skills training for migrant
workers
Perform arts and cultural performances in the
rich cultural life
community of migrant workers
provide high quality compulsory education for the
children of migrant workers
psychological counseling
unions
legal aid
non-government support
voluntary industry
employment information
groups
RISTI, N.º E5, 07/2016 355

education
individual whether or not have professional skills
kinship support physical health condition
children psychological satisfaction
Psychosocial health status (whether or not to feel
the discrimination of urban people, life satisfaction,
geopolitical relations whether with the city people exchanges, etc.)
townsman party
support
have ever received any professional skills training
occupation introduction
Table 5 – The main body of social support

Because of the special status of migrant workers with a large numbers, the issues of
migrant workers have attached great importance to governments at all levels, the
support from government for migrant workers is the strong guarantee for working and
living in cities, especially in enhancing the political status, improving the economic level,
enriching cultural life and other aspects. According to the dual household registration
system, migrant workers cannot enjoy the same education, health care and housing and
other social security. China’s existing social security in the social relief, social welfare,
are almost in a state of legislative blank, just the project of labor and social insurance are
covered by the legislation. In other words, the migrant workers can only rely on their own
meager income to cure themselves if they get sick or injured. The lack of social system
has become the biggest obstacle to the urbanization of migrant workers. the vacancies in
the migrant workers’ medical insurance, endowment insurance cause the difficulty that
the migrant workers cannot enjoy the same treatment like the city people do, because of
the high insurance money and complicated cumbersome procedures, the insured rate of
migrant workers was much lower than the city people. In the Eighth Session of the Fifth
Plenary Session of the Central Committee in October 2015, President Xi talks solutions
about migrant workers’ treatment in the next five years planning. It will improve further
the treatment of workers from the allocation mechanism to migrant workers housing
subsidies aspect. Now governments have nothing to change essentially the gap of the
migrant workers between the city and countryside but solving specific problems and
dealing with negative effects as a manager. Fundamentally, migrant workers citizenization
lies on whether or not governments can create a good system environment, gradually
liberalize restrictions settled, establish a social security system for migrant workers
which mobilize the peasant enthusiasm and improve labor productivity.
Non-governmental organizations generally is non-profit, spontaneity and autonomy,
the characteristics of folk, direct, targeted are shown in the help of migrant workers
from the non-governmental organizations. the problems of migrant workers’ simple
dependence on government and lack of community resource efficiency are made up
with the emerging and developing of non-governmental organizations which effectively
improve the support efficiency of migrant workers. “mixed” phenomenon is caused
because of the lacking of non-government organizations’ legal status. In China, there has
not yet a national non-government organization of migrant workers, and it is difficult to
achieve resource sharing and integration due to a small range and fragmented situation
of a variety of migrant workers. this result is produced because the current law clearly
356 RISTI, N.º E5, 07/2016

stipulates that non-government organization need required acceptance by corresponding

business department. In accordance with the principles of the corresponding hierarchical
management, the establishment of a national migrant workers’ non-government
organization need to find the national ministries and departments which is too difficult.
what’s more, there are other rules that non-governmental organizations with the same
or similar business scope shall not be established in the same administrative region.
Secondly, migrant workers are not legally qualified to organize social organizations with
the “social organization registration management regulations” which migrant workers
can’t get the expression of their interests. In addition, due to the scarcity of resources,
non-government organizations whose income mainly rely on third party funding,
not only have not mandatory tax revenue like the government organizations, but also
generate income through the sale of products and services like business organizations.
Therefore, it is difficult to guarantee the stability of non-government organization by
sense of social justice and warm service.
4.2. Internal Supporting

Migrant workers into the cities basically lose the social relations between the original family
and neighbors, because of rejected by citizens; migrant workers in cities tend to fellow
intercourse. Socially geopolitical relations supporting based on geopolitical relations
of emotions. The mutual supporting in migrant workers plays an important role in the
relationship. 1. Material aspects: mainly financial support and physical support. When
the migrant workers encountered economic difficulties, borrowing money to fellow for
survival generally will be the first choice.2. Information: fellow, as an important source of
information, will take the information (including guidance, advice, sharing information)
from city when they come back. At the same time, they will help farmers who are about
to enter the city to pay or unpaid skills training. 3 psychological aspects: including
spiritual assistance, emotional comfort, etc. Based on the same dialect, similar customs
and similarly psychological identity, migrant workers are easy to understand each other.
Because of lacking social background and lowing social status, economic differentiation
opened the gap between the rich and the poor, political isolation caused class conflicts,
cultural differences gashed perception gap. All aspects of political, economic, cultural and
social limits the scope of interpersonal relationship of migrant workers, which caused the
migrant workers just communicate with fellow or other migrant workers. That reflects
the characteristics of migrant workers’ social relationship of involution. Migrant workers
can only use the direct geopolitical relationship to obtain the basic social support.
Migrant workers as the urban social vulnerable groups, their identity are subjected to
social discrimination, excluded outside the city in a certain extent. They cannot get the
right to meet the needs of social resources and hard to merge with urban residents if
there no effective social support in economy, politics, public service.
Migrant worker are a special kind of social identity that cannot be easily changed with
blood relationship which is obtained by the household registration system. Blood
relationship as the core forms a interpersonal relationship with their parents, brothers
and sisters, children, including the strong relationship between marriage- the spouse.
These people can provide support for the migrant workers themselves with financial
support, psychological counseling, psychological satisfaction, etc. The social relations
based on blood relationship presents the characteristics of supportive and gradually
RISTI, N.º E5, 07/2016 357

sparse. The supportive refers to the formation of strong relationships based on kinship
groups which is not calculated in return and support the greatest efforts for migrant
workers. In addition, because of migrant workers in the city for a long time, they cannot
live together with their parents and children to live together, and the relationships of
the most closed the people are gradually sparse.The social relations of migrant workers
have formed a trend of natural, local, and gradually from homogeneity to heterogeneity.
Nature mainly depends on the relationship between blood and geography, even if two
people have no direct contact, as long as they have a blood relationship or geographical
relationship, they will have the protection of relationship maintenance. In such a
special group of migrant workers, their relationships gradually spread from the blood
relationship to the geographical relationship. Their relationship at the same level is
stronger, they get mutual support and help more easier, then, the cure of the social
status is not conducive to the improvement of social status.
5. Conclusions
In this paper, the system has combed the domestic and foreign economics, sociology,
management and other disciplines related to the floating population of the city into
the relevant literature on the basis of the implementation of the field research. Putting
forward the establishment of the social support system of migrant workers in our
country at this stage is the key to promote the integration of migrant workers into the
city. The social support system can be divided into external support and internal support
according to the main body of the support, the main body of the external support
refers to the government and non-governmental organizations, the subject of internal
support refers to the support based on the geographical relationship and the support
of the blood relationship. At present, the main point of view is that the two element of
household registration system is an obstacle to urban integration of migrant workers,
so the government should abolish this system as soon as possible. This article believes
that migrant workers integrate into the city is a long-term, flow process. First of all, the
two element of the household registration system is based on a certain historical period,
in order to promote economic development and the formation of a set of political and
economic welfare system. Then, the lifting of the system requires the government to
follow up the various supporting measures and the government at all levels of the mutual
cooperation, the introduction of enforceable policy measures. In addition, migrant
workers are just “entering to city”, and to achieve “urban integration”, it is not only has
the city residence, enjoy the urban residents of education, health care and other welfare
benefits, but more importantly, with the psychological and culture integration of urban
residents. So the government in the introduction of migrant workers integration into
the city, not only to pay attention to changes in the household registration of migrant
workers, but also to pay attention to the cultural quality of migrant workers, mental
health and other aspects of the upgrade.
Acknowledgments
This paper is supported by the National Social Science Fund of China in 2014: “Research
on social support network system and policy guidance of peasant workers’ Citizenization
in the process of new-type urbanization”(Serial number: 14BSH0239).
358 RISTI, N.º E5, 07/2016

References
Cheng, Z., Wang, H. (2013). Do neighbourhoods have effects on wages? A study of
migrant workers in urban China. Habitat International, 38, 222–231.
(14), 1–15.
Lu, C., Xu, L. (2012). Sexual and reproductive health status and related knowledge among
female migrant workers in Guangzhou, China: a cross-sectional survey. European
Journal of Obstetrics & Gynecology and Reproductive Biology, 16, 60–65.
Qiu, P., Caine, E. (2011). Depression and associated factors in internal migrant workers
in China. Journal of Affective Disorders, 13, 198–207.
Su,S., Zhang,F. (2014). Factors associated with utilization of reproductive healthcare
services among migrant women workers in Chong Qing, China. International
Journal of Gynecology & Obstetrics, 127, 66–68.
Shen, H., Huang, C. (2012). Domestic migrant workers in China’s hotel industry: An
exploratory study of their life satisfaction and job burnout. International Journal
of Hospitality Management, 31, 1283–1291.
Tao, L., Eddie, C. (2015). Housing choices of migrant workers in China: Beyond the
Hukou perspective. Habitat International, 49, 474–483
Xu, Q., Palmer, N. (2011). Migrant Workers’ Community in China: Relationships
among Social Networks, Life Satisfaction and Political Participation. Psychosocial
Intervention, 20, 281–294.
Zhong,B., Liu, T. (2015). Prevalence and correlates of major depressive disorder among
rural-to-urban migrant workers in Shenzhen, China. Journal of Affective Disorders,
18, 1–9.
RISTI, N.º E5, 07/2016 359

Research on the E-business Enterprise Employee

Occupation Commitment and Career happiness
based on Network Survey
Shuliang Lv1,2, Tiange Duan1, Lanying Du1,*
* lanyingdu@hust.edu.cn
1
School of Management, Huazhong University of Science and Technology, Wuhan 430074, China
2
Zhengzhou Branch China CITIC Bank, Zhengzhou 450001, China
Pages: 360–373
Abstract: E-commerce is a business activity that use of computer technology

and network communications technology. With the development of E-business
platform, Internet competition has changed into the competition of enterprise
talents. This study explores the differences of employee occupation career
happiness in E-business enterprise, As well as the company’s occupation
Commitment under different degree of occupation pressure. Through the design of
an online questionnaire, the study found that the employee occupation happiness
will reduce when facing greater work pressure, Perceived organizational support
played a negatively affect between occupation stress and career happiness. At
the same time E-business employee occupation commitment will be lower when
facing greater pressure, occupation career happiness plays an intermediary role
between occupation commitment and employee occupation pressure. The author
research relationship between occupation stress, employee occupation happiness
and occupation commitment, the result has important implications in reducing
the loss of talents; it is necessary to enhance employee loyalty and cohesion of the
E-business company.
Keywords: Network platform, E-business, career happiness, occupation
commitment
1. Introduction
Because of the fierce market competition, under the crush strength of traction and driven
by endless pursuit of the interests, employees are not easy to the feeling of happiness,
and have great liquidity and lower loyalty. In 2015, the average turnover rate of domestic
employees reached 17.7%, this is regardless of the employees or a company is a very
bad signal. Employees desire to obtain a happy working experience, and the company
is hoping to increase employee loyalty, reduce the flow of personnel in order to improve
the efficiency of production management. With the sense of career happiness of the rise
of the topic, employee career happiness and the causes and effects has been widespread
concern of scholars, has become an important research topic. The personality traits,
360 RISTI, N.º E5, 07/2016

Figure 1 – E-commerce
occupational stress and organizational culture are classified as antecedent variables, and
occupational commitment, work family relations (Qingxiong, 2014).
But the related research is only at the theoretical level, there is no empirical research to
enrich and verify. E-commerce is a business activity that use of computer technology and
network communications technology. With the development of E-business platform,
Internet competition has changed into the competition of enterprise talents.
Figure 2 – E-commerce market

This paper attempts to cut from the occupational stress, research staff occupational
stress and career happiness, professional commitment and perceived organizational
support as occupational stress and career happiness sense of adjustment variables
were measured. Quantitative research results show that the occupational stress more
employees the career well-being is lower, organizational support between occupational
stress and career happiness negative to a significant regulation, organizational support
RISTI, N.º E5, 07/2016 361

Research on the E-business Enterprise Employee Occupation Commitment and Career happiness based on Network Survey
sense is stronger, the occupational stress of career happiness sense of negative to the
weaker the effect of perceived. The greater the occupation pressure of organization staff
occupation commitment is low, occupation career happiness plays an intermediary
role between occupation commitment and employee occupation pressure. Occupation
pressure employees will reduce their own occupation career happiness, so as to reduce
the occupation commitment to the company.
Figure 3 – Professional happiness

Study on this problem both for the occupation career happiness theory supplement and
deepening, or have a very important significance for the management of the company.
First is the theory, with the development of society, as most people want to settle
down in the security is more and more attention and research, but with related work
happiness sense of major concern is short, are engaged in the work to reaction before
work experience of the existing state of influence (Horn, 2004). Kidd (2008) put forward
the concept of occupation career happiness, which is defined as the subjective feeling
of personal occupation career experience and occupation career development process.
This provides clues for us longitudinal studies practitioners working experience. But
because the concept is relatively new, very lack of empirical research, this study enriches
the empirical research on the occupation career happiness, and provides clues for the
study. In practice, this study provides a way of staff management, including how to set
appropriate working pressure, how to improve employee well-being and how to improve
the occupation commitment of employees, improve the management efficiency.
2. Theoretical background and hypothesis

2.1. Occupation pressure
Occupation pressure is when the occupation requirements caused by forcing people to
deviate from the normal function changes the pressure. It is also called the occupation
362 RISTI, N.º E5, 07/2016

pressure, is a multidimensional concept rather than a single dimension (Lin,2003). A

variety of reasons of the occupation pressure of diversity, more focused individual -
environment interaction theory, individual environment matching theory, control theory,
job demand (Lazarus,l978; Caplan, 1980; González, 2015). Individual environment
interaction theory is the core of occupational stress as a result of combined effects of
individual and environment, and individual environment fit theory that occupational
stress depends on the ability of the individual and the environment matching degree,
when individuals ability to adapt to the environment, occupational stress will is
relatively low, and their ability and the environment does not match, it will produce a
higher occupational stress. The job demand control theory that occupational stress by
two effects, the working environment of job demands and job control. High demand and
low job control human occupational stress most. Tremendous professional pressure will
seriously affect their self-confidence and enthusiasm, easily lead to burnout (Karasek,
1979), employees in order to cope with the pressure of the high occupation, will consume
more personal time and effort to accomplish the task, so that job satisfaction will also
fall (Xiaoming, 2004), in addition to this, there are many other negative results we will
not enumerate.
2.2. Occupation pressure, perceived organizational support and

career happiness
Career happiness of late, Kidd (2008) will be the, and highlight it with the happy sense
of difference is that career happiness sense not only refers to the now engaged in the
work of the feelings, but also including the career began from the first job of feelings and
experiences. The concept to today’s social environment more fit, with the mobility of
employees of economic development is becoming more and more strong, many people
will to high frequency continuous change of the work, this will lead to their feeling
invisible of the existing state would be affected by the prior experience of. Therefore, the
career happiness sense the longitudinal development concept as a measure of employee
status indicators more reference value.
The rapid development of society and economy must be accompanied by high intensity of
market competition. Enterprises in order to maximize the win the living space to obtain the
long-term development, are often applied to staff a strong occupational stress, may be the
body of high labor strength, high-risk operations, there may be psychological indicators
of performance appraisal. The empirical study found that employees will seriously affect
the mental health in the long term high pressure state, emotional self-regulation function
decline, irritable and anxious (Lu, 1997; Liang, 2005). Therefore happiness decreased
significantly. Some employees are under too much pressure, edge on the brink of collapse
of body and mind, and for the physical and psychological self-protection mechanism,
especially easy to slack, escape from work and job burnout. Therefore, we can infer that
between employee occupation pressure and occupation career happiness is negative
correlation. Eisenberger (1986) put forward the concept of perceived organizational
support, the staff of the organization to describe the importance of their contribution, the
overall feeling and concern for their welfare. Perceived organizational support can have a
RISTI, N.º E5, 07/2016 363

positive impact on the behavior of employees, so that employees exhibit more moral and
civic behavior. Therefore, we have reason to believe that the perceived organizational
support can alleviate the psychological pressure of employees because of discomfort; get
more psychological comfort, so as to enhance their occupation career happiness. In view
of this the author put forward the following hypothesis:
Hypothesis 1: Occupation pressure greater employee occupation career happiness
is weak.
Hypothesis 2: When perceived organizational support is high, the effect of employee
occupation pressure on the occupation career happiness will reduce.
2.3. Occupation pressure and commitment

Occupation commitment refers to rely on personal or professional occupation of identity
and emotion, which led to the occupation or professional investment and internalization
of social norms to change occupation degree (Jepson, 2006). Under great pressure
due to employee occupation, prone to depression and lower happiness, loyalty to the
organization will be relatively low, difficult to make high occupation commitment.
Studies show that the occupation pressure will affect the Occupation Commitment
of employees, employee occupation pressure showed lower levels of occupation
commitment (Eisenberger, 2002). Through the study found that a good mental state and
positive emotion of occupation commitment has a positive role in promoting. Higher
employee occupation career satisfaction has low occupation pressure, the psychological
state is positive; it is easier to make occupation on organizational commitment, high
loyalty (Farh, 2007). On the contrary, engage in occupational stress under the drive of
employee exhausted, career happiness is very low, is easy to produce slack and escape
from the idea, so occupational commitment will is relatively low. Based on the above
analysis, we can draw the following assumptions:
Hypothesis 3: The greater the high pressure of employee occupation, occupation
commitment will be lower.
Hypothesis 4: Occupation career happiness plays a mediating role between occupation
stress and occupation commitment.
2.4. Conceptual model

The relationship between mode of occupation pressure, occupation career happiness
and Occupation Commitment Based, this study proposes an integrated conceptual
framework. Occupation pressure greater employee occupation career happiness is
low. The perceived organizational support is more high, occupation pressure on the
occupation career happiness weakened. Occupation career happiness is higher, the
higher the occupation commitment to customers. Occupation career happiness plays a
mediating role between occupation stress and occupation commitment, the concept of
concrete framework was shown as figure 1.
364 RISTI, N.º E5, 07/2016

Figure 4 – The research framework
3. Data analysis and discussion

3.1. Sample selection and characteristics
Through an online questionnaire (Sojump online questionnaire) way to collect
data, we selected 268 practitioners from 20 different industries as research
subjects, minimizing industry bias. To make active and serious subjects involved
in the investigation, as a sample of two three prize rewards the lucky participants.
Questionnaire comprehensive understand occupational stress, organizational
support, career happiness and career commitment of information. Before the
questionnaire official launch, we make pre-survey by choosing 50 research subjects,
then adjustments Q items based on feedback. Questions of the questionnaire are
using the traditional five-point Likert scale to measure.
In this study, the survey recovered 252 questionnaires; the analysis excluded incomplete
questionnaires answered 14 parts. Sex ratio questionnaire respondents are: women
accounted for 42.6%, men accounted for 57.4%; the age structure: age under 30 years
accounted for 34.5%, 30-50 years old accounted for 46.2%, 19.3% over 50 years;
educational structure is: undergraduate accounted for 56.7%, college accounting for
23.7%, the remaining 19.6%; monthly family income structure (RMB) is: 2000-5000
accounted for 27.4%, 5000-10000accounting for 42.1% , 10000-30000 accounted for
28.5%, the remaining 2%.
3.2. Variable measurement

Measurement of the occupation pressure, general research is the use of foreign mature
scale. This is the occupation stress scale developed by Cooper et al. The scale consists of
six dimensions, job characteristics, role stress, interpersonal relationship, occupation
development, organization structure and family work conflict. It will be the concept
of effective occupation pressure quantization and structured, has been recognized
by scholars. At the same time, according to the specific circumstances of the slightly
modified, and ultimately the formation of 22 items of occupational stress scale, including:
your work is complex, workload is big; you do not finish the work, basic no time to rest;
task with difficulty and challenge.
On the measurement of organizational support, Eisenberger (1986) design the 36
items of perceived organizational support measurement table, but because too many
questions, respondents are prone to fatigue and boredom, is not conducive to the
RISTI, N.º E5, 07/2016 365

collection of objective and effective data. Farh (2007) simplified Eisenberger scale,
and we also choose this simplified version of the questionnaire, the specific translation
reference Doctoral Dissertation on organizational support measurement items (weiwei,
2014). Specific topics include: the company will consider my opinion; the company does
take into account my welfare; the company will consider my personal goals and values;
when I have difficulties, the company will help me; if I was good and do wrong, the
company will forgive me etc..
Because the occupation career happiness of the concept put forward later, so the relevant
research is very few, the measurement method of occupation career happiness discussion
and study less. At present, only Creed (2013) in their recent study by measuring the
occupation satisfaction and occupation problems to evaluate the occupation career
happiness. Career satisfaction measurement was construction by using Minnesota
Satisfaction of type long table (MSQ), which a total of 21 questions, including the ability
to keep busy. From time to time to have a chance to do some different things; in the
group become important role; colleagues get along; perform well on the job, the reward
is required. Occupation distress as measured by Lisa (1994), who designs occupation
distress scale. Items include: I often in the choice of profession or occupation feel
frustrated, or depressed. I often because the choice of a profession or occupation of those
did or didn’t do things to blame yourself; I often feel my life without a lot of purpose.
The occupation commitment measurement was developed by Blau, which has 7
items(Fornell, 1981;Yujing, 2015). Translation scale Likert scale, subject: my career is
very good, I will never give it up, even if there is a better treatment of the new career
opportunities appear, I will not give up now career; if I were to do a choice, I also
engaged in the present job; I would very much like to now in this profession to develop
their own; even if you don’t need to work there is enough wealth, I will still career in
the now; I had choose this career is happy; I am engaged in the occupation ideal, it is
worth I entrust lifelong.
4.1. Analysis of reliability and validity
This study use SPSS 19.0 to the questionnaire reliability and validity analysis,
reliability using Cronbach’s a coefficient, exploratory factor analysis and confirmatory
factor analysis of the construct validity of the questionnaire test, exploratory factor
analysis using principal component analysis method, orthogonal test and variance
maximization rotation method extracting common factor. Table 1 shows all of the
variables of the measurement items, and the factor loading value. Table 1 shows,
the Cronbach’s a coefficient of each factor between 0.786 and 0.907, is greater than
all of the acceptable level of 0.700, indicating that all variables in the questionnaire
measurement items with high internal consistency, which have higher reliability.
From table 1 can be seen, the variables of the measurement items of factor loading in
0.531 and 0.971 and the factor loading of the majority of projects are above 0.700, no
multiple load. It is proved that the scale has good discrimination validity.
366 RISTI, N.º E5, 07/2016

Load Cronbach’s
variable Measurement items
factor a
1.your work is complex and large workload .746
2.you always have to do the work .884
3.work tasks is difficult and challenging .630
4.you often work overtime .795
5.you do not understand the scope .757
6.your work without a clear explanation .538
7.Sometimes different job requirements .669
8..sometimes assigned to different areas .531
9.conflict with colleagues or unpleasant .649
10.lack of support .625
11.rarely able to get sympathy and help .805
Occupational 12.work feel isolated .634

.836
Stress
13.slow or no promotion .795
14.not enough space for personal development .874
15.often occurs contradiction between work and
.719
learning, training
16.units of the organization structure is rigid .762
17.lack of effective communication between
.631
departments
18.compensation system is not reasonable .859
19.much work unit level, lack of efficiency .638
20.unreasonable regulations and procedures .742
21.understanding and support is not enough .765
22.take care of the family, you need to invest a lot of
.865
energy
1.company will consider my opinion .609
2.company is really regard for my welfare .804
3.company will consider my personal goals .828
Perceived 4.when I have difficulties, the company will help me .741

Organizational 5.if I do something wrong, the company will forgive me .754 .887
Support
6.if there is a chance that the company will use the me
.868
(R)
7.companies are very care for me .730
8.if I have special needs, the company will give help .707
RISTI, N.º E5, 07/2016 367

1.keep busy in work .607

2.opportunity to work independently .750
3.do some different things to chance .789
4.become an important role opportunity .608
5.my boss treats his / her subordinates way .850
6.my boss to make a decision .821
7.able to do some things not contrary to my conscience .826
8.stability of my work .759
9.do something for other people .770
Occupation 10.tell others what to do with the opportunity .706
.807
satisfaction 11.I can give full play to the opportunity .594
12.implementation of company policies .644
13.my income and my workload matching .851
14.job promotion opportunities .886
15.make their own judgments of freedom .733
16.decide how to complete the work .796
17.working conditions .632
18.way to get along with colleagues .709
19.excellent performance, rewards .778
20.I can get from the work achievement .873
1.I often choose occupation when frustrated (R) .827
2.I often feel helpless in choosing occupation career
.777
planning (R)
3.I tend to avoid or delay, until the last moment (R) .730
4.I think I should do as soon as possible but not to do,
.800
it makes me feel anxiety (R)
5.I choose a satisfactory profession or occupation will
.709
feel tension and pressure (R)
6.I often blame myself because in those who did or
.833
didn’t do anything (R)
7.I tried to eliminate any irresolute and hesitant,
Occupation .861
pretend it doesn’t exist (R) 0.786
problems
8.I often feel my life is a lot less purpose (R) .736
9.I often wish I could disappear in the selection of all
.786
the problems (R)
10.I have no special talent to follow the first occupation
.875
choice I (R)
11.influential person does not agree with my
.969
occupation choice will hinder me (R)
12.every day I will take the time to think about the
.699
problem, and how can I do (R)
13.I considered discipline, I seem to get a lot of
.650
contradictory or confusing information (R)
368 RISTI, N.º E5, 07/2016

my occupation is very good .701

I will not give up the occupation .623
if I have to do a choice, I will now engaged in
.905
occupation
Occupation I am very willing to develop their own occupation in
.827 0.898
Commitment this field
Even this work not provide enough wealth, I will still
.711
be engaged in the occupation
I had to choose this occupation is happy .787
I now engaged in the occupation ideal .971
Table 1 – Exploratory factor analysis and reliability index
Table 2 is a latent variable correlation coefficient table, the Pearson correlation

coefficient between the main measurement variables, analyze the results, show that the
occupational stress and organizational support and career happiness sense of significant
correlation, occupational stress and professional commitment is significantly related to,
for verification of occupational stress, career happiness sense of relationship and career
commitment provide preliminary evidence.
occupational organizational Occupation career

stress support Commitment happiness
Occupational stress 1
Organizational support .070 1 .
Occupation Commitment -.356** .050 1
Occupation happiness -.488** .421** .589** 1
Mean value 3.3212 3.1560 2.3640 2.4315
Standard deviation .65631 .48895 .92121 .87675
AVE 0.643 0.684 0.753 0.787
Table 2 – Correlation coefficients of latent variables

In addition, we also examined the validity of the measurement, measuring whether
there is confusion between the variables. As shown in Table 2, four latent variables in
this study measured the average variance extracted (AVE) were greater than the square
it with the correlation coefficient between the other variables, suggesting that the
measured variables have good discriminant validity (20).
RISTI, N.º E5, 07/2016 369

4.2. Model test

1. The mediation effect of inspection
Occupational stress Occupation happiness Occupation Commitment

Independent
variables β β β
(Standardized t-value (Standardized t-value (Standardized t-value
Coefficients) Coefficients) Coefficients)
C 3.137*** 11.466 3.436*** 21.367 -.3512 -.7481
Occupational
-0.638*** -3.837 -0.422*** -2.883 0.016 0.045
stress
Occupation
0.451*** 4.153
happiness
Adjusted R2 0.482 0.456 0.675
R2 0.576
F 6.845*** 11.674*** 16.754***
df 1 1 2
N 130 130 130
Table 3 – The mediating effect of risk propensity

By the analysis of data table 3 can be seen, occupation pressure and occupation
commitment negatively related to occupation career happiness plays an intermediary
role between occupation stress and occupation commitment, Hypothesis 3 and
hypothesis 4 was passed.
2. Regulation Effect Test
Occupational stress Occupation happiness

Independent variables
β (Standardized β (Standardized
t-value t-value
Coefficients) Coefficients)
C 1.169*** 2.796 -1.487 -.742
Occupational stress -.506*** -4.283 -.213 -.312
Perceived organizational
.314*** 3.816 .530*** 3.343
support
Occupational stress *
Perceived organizational -.217** 1.940
support
Adjusted R2 0.648 0.789
F 15.156*** 154.546***
df 2 3
N 130 130
Table 4 – Test the moderating effect of professional investment
370 RISTI, N.º E5, 07/2016

By the analysis of data table 4 can be seen in the sense of organizational support between
occupation stress and occupation career happiness plays a negative role. The greater
the pressure of employee occupation, occupation career happiness is weak; employees
perceived organizational support is strong, occupation pressure on the occupation career
happiness and negative effect is weak, Hypothesis 3 and hypothesis 4 was passed..
5. References
This article attempts from the beginning of occupational stress, research staff
occupational stress and career happiness, professional commitment and perceived
organizational support as occupational stress and career happiness sense of adjustment
variables were measured. Quantitative research results show that the occupational
stress more employees the career well-being is lower, organizational support between
occupational stress and career happiness negative to a significant regulation,
organizational support sense is stronger, the occupational stress of career happiness
sense of negative to the weaker the effect of perceived. The greater the pressure of
employee occupation commitment is lower, occupation career happiness plays an
intermediary role between occupation commitment and employee occupation pressure.
Occupation pressure employees will reduce their own occupation career happiness, so
as to reduce the occupation commitment to the company. May be due to occupational
stress of employee’s long suffering to physical or mental pressure destruction, emotion
regulation function decline, prone to irritability or anxiety and other negative emotional
state, to work to aversion to hide such attitude, so career happiness sense is very low.
Perceived organizational support more employees can get energy from the collective,
effectively regulate the pressure work of discomfort, which has relatively high occupation
career happiness. Occupation pressure staff is relatively difficult to obtain success in
the job and satisfaction, and love to work is relatively low, it is difficult to maintain
high loyalty, occupation commitment. If the employee’s occupation career happiness is
high, which can effectively alleviate the discomfort brought occupation pressure, thus
weakening the negative relation.
The main theoretical contributions of this paper are: through empirical research
demonstrates the relationship between occupation stress, organizational support,
occupation career happiness and organizational commitment. The related research
enriches the antecedents and consequences of this emerging occupation career happiness
concept. Practice, the study of employee management provides ideas to moderate
working pressure, and given sufficient organizational support, let employees feel warm;
recruit positive psychological quality good employees, these employees career happiness
are relatively strong, can make a higher career commitment, on the organization of the
higher loyalty. In addition, this study also has some limitations. First, this study career
happiness sense is measured by combined measurements of the professional satisfaction
and troubled two related concepts, now is not specialized mature career happiness
feeling questionnaire for research use, therefore results inevitably biased, the future can
be considered in the development of a professional scale. Secondly, the questionnaire
is mainly concentrated in the area of Henan Province, to other areas of Henan province
and is favorable to explore, the future can be considered to further expand the scope of
the study and the research sample.
RISTI, N.º E5, 07/2016 371

References
Caplan, R., Cobb, S. (1980). Job demands and worker health: Main effects and
occupational differences. Ann Arbor, MI: Institute for Social Research.
Creed, P. (2013). Compromise, well-being, and action behaviors in young adults in
career transition. Journal of Career Assessment, 21, 3–19.
Eisenberger, R. (1986). Perceived Organizational Support. Journal of Applied
Psychology, 71, 500–507.
Eisenberger, R., Huntington, R. (2002). Organizational Support. Journal of Applied
Farh, J., Hackett, R. (2007). Individual-level cultural values as moderators of perceived
Organizational support-employee outcome relationships in China: Comparing the
effects of power distance and traditionality. Academy of Management Journal, 50,
715–729.
Fornell, C. (1981). E valuating Structural Models with Unobservable Variables and
Measurement Error. Journal of Marketing Research, 18, 39–50.
(14), 1–15.
Horn, J, (2004). The structure of occupational well-being: A study among Dutch teachers
.Journal of Occupational and Organizational Psychology, 77, 365–375.
Jepson, E., Forrest, S. (2006). Individual contributory in teacher stress. The role of
achievement striving and occupational commitment. British Journal of Educational
Kidd, J. (2008). Exploring the components of career well-being and the emotions
associated with significant career experiences. Journal of Career Development, 35,
166–186.
Karasek, R. (1979). Job demands, job decision latitude, and mental strain: Implicatios
for job redesign. Administrative Science Quarterly, 24, 285–308.
Lisa, M., Alan, L. (1994). The Development and Validation of Coping with Career
Indecision. Journal of Career Assessment, 2, 91–110.
Lin, S. (2003). Current situation and direction of the working pressure. Psychological
science, 20, 494–497.
Lazarus, R. (l978). Stress-related transactions between Person and environment. In
L.A.Pervin & M.Lewis(Eds.),perspectives in international psychology, New York:
Plenurn.
Lu, L. (1997). Occupational stress in Clinical nurses. Counseling Psychology Quarterly,
1, 39–50.
372 RISTI, N.º E5, 07/2016

Liang, S., Hsieh , A. (2005). Individuals perception of career development and job
burnout among flight attendants in Taiwan. International Journal of Aviation
Lirong, L. (2000). The theory and measurement, Occupation Commitment. Psychology
dynamic, 8, 39–45.
Qingxiong, W., Yin, C. (2014). Occupation career happiness concept introduction,
theoretical framework and research prospect. Foreign economy and management,
36, 56–62.
Weiwei, J. (2014). Employees perceived organizational support, study the relationship
between psychological ownership and voice behavior. Jilin University doctoral
dissertation, 20–24.
Xiaoming, L. (2004). Occupation pressure, teaching efficacy,relationship between
the primary and secondary school teachers occupation burnout. Psychological
development and education, 2, 56–61.
Yujing, P. (2015). The knowledge staff occupation, Occupation Commitment and job
attitude called relationship research. Management science, 28, 103–114.
RISTI, N.º E5, 07/2016 373

The Influence of Rural Financial Development on

Urban-rural Income Gap based on Data Mining
Method
Qiwen Zhang, Weiwei Lu, Jinyuan Wang*
* 5390748@qq.com
Pages: 374–386
Abstract: In this paper, the authors analyze the income gap between urban and
rural areas from the perspective of rural financial development. By using data
mining method, we construct the rural financial development indicators, and then
make empirical analysis based on vector autoregressive (VAR) model. At the same
time, using panel data model to study the influence of rural financial development
in 30 provinces of China. The result shows that the scale and efficiency of rural
financial development can narrow the income gap between urban and rural areas.
The impact coefficient of rural financial scale is 0.019, and the significant level is 10%,
which showed that the rural financial development scale increased by 1 percentage
points will lead the income gap reduce 1.9%. Also, we found that there are obvious
differences in different regions; the efficiency of rural financial development in the
eastern and western regions plays a negative role, while in the central part plays a
positive role.
Keywords: Data mining, rural finance, income gap, vector error correction
1. Introduction
Since the reform and opening up, the continuous expansion of urban and rural areas is
regarded as a potential threat to the development of social harmony is not conducive
to the harmonious development of urban and rural areas is the main embodiment of
the income gap between urban and rural residents. The relative ratio of urban and
rural residents’ income has expanded year by year, which has exceeded 3: 1. Due to
the complex causes of the income gap between urban and rural areas, most of that for
the following reasons: the beginning of reform and opening up agriculture product
price reasons, urban and rural residents’ boundaries and hinders the farmers to
the household registration system, and the government implemented the policy of
urbanization and other reasons (Clark, 2006; González, 2015). In recent years, some
scholars have studied the relationship between financial development and the income
gap between urban and rural areas, but there are few independent from the perspective
of rural financial development to explain the reasons for the income gap between urban
and rural areas (Levine, 2000; Townsend, 2006; Chen, 2007). The from the perspective
of the development of rural financial research its impact on the income gap between
374 RISTI, N.º E5, 07/2016

urban and rural areas, to rich the income gap between urban and rural areas of Omni-
directional and multi-level understanding has important theoretical significance,
to alleviate the problem of income gap between urban and rural areas in our country to
construct harmonious society has a positive and practical significance.
This article will start from the rural financial development to the income gap between
urban and rural areas, the use of theoretical analysis and empirical analysis to examine
the relationship between the two. In this paper, we first use the comparison and statistical
analysis of index system to describe and describe the rural financial development
and urban-rural income disparity in different regions, and then give the normative
conclusions (Li, 2007; Chen, 2009). The subsequent theoretical analysis, by promoting
economic growth, income distribution and threshold effects and other financial factors
analysis of rural financial development on the income gap between urban and rural areas
the influence mechanism and draw the theoretical results, and the theoretical analysis
based on empirical analysis. Empirical analysis will based on vector auto regression
model, and on this basis, the Johansen cointegration test to determine variables between
whether there is a long-term equilibrium relationship between timing model constructs,
but this only that there is a relationship between the variables, and not that direction of
the causal relationship of the variables, then using Granger causality test to determine
the causal relationship between the variables. Then establish correction error vector
(VEC) model to determine the short-term balance relationship between the variables
and then use the impulse response function and variance decomposition to examine the
explanatory variables influence of explanatory variables and the degree. In addition, the
paper will build a panel data model to study the impact of rural financial development
in 30 provinces and cities in China, not only from the vertical angle but also from the
perspective of horizontal analysis of the relationship between the two, the method is
more comprehensive. Combined with time series theory data model and panel data
model, the results obtained are more reliable, more scientific and more comprehensive.
2. Data mining theory

Data mining is a large number of relevant data base on the data to explore and establish
a model of advanced methods, the data source if the massive, real, and real data most of
all there is noisy or missing; users are interested in and required knowledge is certainly
can be mined; the discovered knowledge for decision maker is service, so it is acceptable,
easy to understand, can be widely applied, and can solve some of the problems found.
The research of data mining technology is integrated with the technology of multi subject
intersection, which makes the data mining technology presents a variety of forms. Any
method has its application field, at the same time; any kind of method is not a panacea, it
is impossible to solve all kinds of problems. Mature data mining system should integrate
many kinds of mining methods; analyze data from many angles, in order to make up the
deficiency of single data mining method.
•• Artificial neural network method: Neural network in recent years by
the people’s attention, due to its own has good robustness, self-organization,
adaptive, parallel processing, distributed storage and high fault tolerance
and other characteristics, processing complexity problem provides relatively
effective and simple method. It is modeled on the structure and function of
RISTI, N.º E5, 07/2016 375

The Influence of Rural Financial Development on Urban-rural Income Gap based on Data Mining Method
Figure 1 –Artificial neural network
human neurons, which can be used to generate patterns for prediction and
classification. It establishes three kinds of neural network models, which are
based on the MP model and the Hebb learning rule.
•• Cluster analysis: Clustering analysis is mainly study in accordance with the
“Like attracts like.” thought, is a non-supervised learning method. To the large
set of data points designated cut into several groups (or classes), makes the data
in each group between the high degrees of similar, the distance between possible
small; and in different group data with a high degree of dissimilarity between
distances as far as possible to large. Clustering analysis method is widely used
in the field of classification, clustering in biology through the derivation of the
classification of plants and animals, according to the similarity function of
genetic gene classification, so as to get the inherent structure of the population.
In business, clustering can help market analysts to find some rules from the base
of the consumer buying patterns.
•• Decision tree method: The decision tree method is a common prediction
modeling method for classification, clustering and prediction, it mainly uses the
idea of divide and conquers, and the first problem search space is divided into
several parts, and then is processed sequentially. Therefore, the decision tree
to establish the growth process is actually the process of data rules generation.
Decision tree method is simple and fast, it is very suitable to deal with non-
numerical data, it is easy to visualize the data into a rule.
•• Association rules: The main technology of the association analysis is the
association rules, which is one of the main research fields of data mining and
one of the mature technologies. It can identify the relationship between the
data items in a given data set and structure characteristics, and describe the
relationship between the data items. In the retail industry and related technology
can help the operator found that consumers buy hobby, an accurate description
of characteristics of consumers and to provide consumers with the purchase;
can help to analyze the consumer cycle time, reasonable arrangements for
inventory and purchase.
Data mining generally through data preparation, data mining, the results of expression
and interpretation of the three stages, it is a forward, and complete the learning process.
The following figure 2 shows data mining process shows the process of mining previously
unknown, valid, and available information from the mass of data to form the knowledge.
376 RISTI, N.º E5, 07/2016

Figure 2 – Data mining process
The basic idea of ID3 algorithm is that the attribute selection metric is used to select the
best attribute as the node. The greater the value of the information gain, the smaller the
uncertainty, so that when the test of each non leaf node, can get the maximum number
of categories of information being tested. The desired information for the classification
of a given training data is given by the following formula:
m
Info ( D ) = − ∑ p log ( p )
i =1
i 2 i
(1)
Suppose that the tuple in D is divided into A according to the attribute V, in which the
attribute A can be measured by the following formula, according to the observation of
the training data:
v Dj
Info ( A ) = ∑ D
( )
Info D j (2)
j =1
The information gain to be obtained from the branch on the attribute A can be
described as:
Gain ( A ) = Info ( D ) − Info ( A ) (3)
ID3 algorithm has its advantages as follows: the basic principle of the algorithm is clear;
the classification speed is fast; the practical value of the sample learning algorithm.
C4. 5 algorithm is an extension of ID3 algorithm, an early machine learning algorithm,
it is also a common algorithm to construct decision tree classifier, which is the basis
of many decision tree algorithms. Information gain rate (ratio gain) as an extension
of the information gain, to solve the problem of ID3 shortcomings. The segmentation
information is defined as follows:
y D  Dj 
SplitInfo ( A ) = − ∑
j
× log 2   (4)
j =1 D  D 
 
RISTI, N.º E5, 07/2016 377

Information gain ratio equals the ratio of information gain to segment information:
GainRatio ( A ) = Gain ( A ) SplitInfo ( A ) (5)
Classification and regression tree is a technique to create a binary decision tree, cart
algorithm to Gini coefficient and variance based selection optimal grouping variables
and split point, from selected Gini coefficient is the minimum value of the property as
the test attribute, if the smaller the Gini coefficient, divided more reasonable, purity of
the sample set higher. If the training tuple set D contains a record of the M category,
then the Gini refers to the calibration:
m
Gini ( D ) = 1 − ∑p
i =1
2
i (6)
If D is divided into two parts, D1 and D2, the Gini coefficient of the division is
D1 D2
Gini ( D ) =
D
( )
Gini D1 +
D
( )
Gini D2 (7)
3. The influence mechanism of rural finance

The role of finance is to optimize the allocation of funds, financial institutions to absorb
the savings, to fund the demand for loans to invest; the ability to convert deposits into
investment is the financial efficiency. The development of rural finance is directly related
to the formation of the original capital and the optimization of capital allocation. The
first increase in rural primitive accumulation of capital is the most basic problem of
total savings, also is to expand the scale of finance, because rural residents savings can
contribute to rural capital, which laid the foundation for the rural construction funds,
there is conducive to technological progress and innovation in rural areas, promote the
rapid development of the rural economy, increase farmers’ income and the income gap
between urban and rural areas; and if the capital into investment, which is the financial
efficiency, the proportion reflects the social reproduction ability, high conversion rate
will lead to the expansion of social reproduction, is the improvement of the economic
efficiency, so as to achieve economic growth, increase income, reduce the urban-rural
income gap. To expand the scale of the rural finance, and transformed into investment,
improve financial efficiency, make the capital can get effective configuration, the way
farmers can be through direct investment to increase its monetary income, and secondly
investment can also promote the development of rural economy, improve the living
standards of rural residents, and indirectly, to increase the income of peasants, and is
conducive to narrowing the income gap between urban and rural areas. In addition,
financial development promotes the economic growth after stimulated the high income,
investment demand, raising interest rates, low income people can also through the
provision of loans and higher interest income, which will be to narrow the income gap
between urban and rural areas.
378 RISTI, N.º E5, 07/2016

The income distribution system implemented in our country at present stage is according
to the labor distribution as the main body, the various modes of distribution coexist,
and all kinds of production factors participate in the distribution according to the
contribution. Production factors mainly include labor, land, capital and entrepreneurial
talent, the difference in the endowment of individuals can lead to serious income gap.
In the countryside, people are educational opportunities is far less than that of urban
residents, China’s fifth census only in Zhejiang Province, the results from the point
of view, rural and urban population aged 15 and above the average years of schooling of
respectively 6.58 and 850 years, a difference of nearly 2 years, the same age rural and
urban population illiterate rate of 11.13% and 5.95%, respectively, so the rural population
relative to urban population in terms of the lack of entrepreneurship. Again due to
the rural population occupies the initial wealth level is low, mainly rely on labor and
land, the two factors of production to obtain income, and our country land is national
and the collective common all, and is not a personal owned private wealth, cannot
arbitrarily transfer, so the land does not as farmers increased wealth of resources, so
the production according to the contribution of production factors participate in the
distribution of the urban - rural income produces a huge difference. Of the four elements,
capital is the core of the financial resources, capital to enjoy the financial services, which
resulted in the difference of the residents in urban and rural areas of financial services
can be acquired, and the rural residents lack of financial resources to rural financial
development efficiency and scale expansion was restrained, which is owned by urban
and rural residents’ financial resources, farmers do not enjoy the income wealth created
by the financial resources, widen the income gap between urban and rural areas.
Secondly, finance can change the use of structure and distribution of social resources,
access to financial support more departments will promote the improvement of the
efficiency of industrial production, change the industrial income distribution, resulting
in the income differences between individuals. In the rural financial market, most
of the rural capital is lost through financial institutions, which is configured to non-
agricultural industries, resulting in the problem of financial vacuum in agriculture and
the development of rural finance. Financial institutions will return on investment rate
higher departments there is a deviation, on the low productivity of agriculture set the
“threshold”, a serious preference for high production efficiency of industrial economy,
so the rural financial institutions have large capital outflows, the phenomenon is due
to reach the “threshold” and the lack of innovation and development, rural economic
growth has been hampered by the Department of financial institutions to set up the
“threshold”, agricultural productivity is not high, the income gap between urban and
rural areas is further widened.
4.1. Index selection
1. Urban and rural income gap index (IG): In this research, the time
series model is used to measure the ratio of per capita net income of urban
residents and rural households as a measure of the income gap between urban
and rural residents.
RISTI, N.º E5, 07/2016 379

2. Rural financial development index: (1) the development scale of the rural
finance index (rfir), a measure of indicators of financial development scale has
been widely accepted as the Malmquist index and GE’s index, McIntosh index
is the ratio of M2 to GDP, China’s financial system is bank oriented, and the
main business of the bank is lending and deposit. Therefore, this paper with
rural deposit loan balance and and agricultural GDP ratio as the scale of the
development of rural financial indicators. (2) the rural financial efficiency index
(RFE), which measures the efficiency of the financial intermediaries into the
investment loans, so this paper used the ratio of rural loans and deposits to
measure. The ratio reflects the efficiency of the rural economy to obtain loans,
3. Control variables: (1) urbanization index (urb), with the urban population
accounted for total population proportion as the city rate, the income gap
between urban and rural areas affected by the impact of urbanization level, the
improvement of the level of urbanization will significantly narrowing the income
gap between urban and rural areas; (2) fiscal expenditure index (FSA), with
spending on agriculture accounted for the total fiscal expenditure proportion
said. Chinese government intervention on the change of the income gap between
urban and rural areas has an important influence, and government intervention
is government expenditure, expenditure for agriculture accounted for total fiscal
expenditure directly proportional relationship between the rural economic
development, the income of rural residents; (3) rural fixed assets investment
index (RI), with total investment of fixed assets in rural areas accounted for
the whole social fixed assets investment proportion, infrastructure construction
will promote economic development in a certain area, will lead to the growth of
a series of local enterprises, leads to an increase in people’s income, life quality
promotion, to narrow the income gap between urban and rural areas has a
crucial role.
4. Panel model index: Panel data model selection index and the time series
data in the theoretical model of the selected index is consistent, only the data
into regions: the income gap between urban and rural areas (IG*) is the area
of urban households disposable income and rural households per capita net
income ratio; rural financial development scale index RFIR* and efficiency
index RFE* is rural deposits and loans and agricultural GDP
4.2. Model setting

The timing model in this study the changes of income gap between urban and rural areas
from the perspective of rural financial development research, the income gap between
urban and rural areas (Ig), the development scale of the rural finance (rfir) and rural
financial efficiency (RFE) set for the endogenous variables, the fiscal expenditure on
supporting agriculture (FSA), rural investment in fixed assets (RI) and urbanization
(urb) as an exogenous variable, based on the AIC and SC criteria model lag order number
is 2, VAR model:
IG = α 1 RFIRt −1 + α 2 RFIRt −2 + α 3 RFEt −1 + α 4 RFEt −2 + α 5 FSA + α 6 RI + α 7URB + c + ε
380 RISTI, N.º E5, 07/2016

And in the VAR model based on further establish the vector error correction model
(VEC) model with VAR cointegration constraints model, the research is variable between
short-term fluctuation effects, there is a co integration relationship between variables. If
the lag period of VAR is set to K, then the lag period of VEC is k-1, so the VEC lag order
is 1, and the specific expression is as follows:
DIG = α 1 ECMt −1 + α 2 DIGt −1 + α 3 DRFIRt −1 + α 4 DFSA + α 5 DRI + α 6 DURB + ε
Panel data model to test effect of the various regions of the rural financial development
scale, the efficiency of rural financial development, urbanization level, and fiscal
expenditure for agriculture and rural fixed assets investment on local variations in the
income gap between urban and rural areas, the following model is established.
IGit* = α i + β1 RFIRit* + β2 RFEit* + β 3URBit* + β 4 FSAit* + β 5 RI it* + µit
4.3. Model test

In this paper, the SIC criterion is used to determine the lag order. The IG, RFIR, RFE,
FSA, URB and RI and their first order sequences are tested by ADF, and the final results
are shown in table 5.1. From table 5.1, we conclude that the variable Ig, fir, Fe, CZ, RI and
urb sequence under 10%, 5% and 1% significant level were not significant, that they are
nonstationary time series, and their first order differential sequence are stationary, so a
level value of each variable are first-order single whole time series I(1).
variable (C,T,L) ADF test value critical value result
IG (C,N,1) -1.246118 -2.627420 non- stable

DIG (C,N,0) -3.735636*** -2.627420 stable
RFIR (C,N,0) -0.226843 -2.625121 non- stable
D RFIR (C,N,0) -3.921127*** -2.627420 stable
RFE (C,N,0) -1.568039 -2.625121 non- stable
DRFE (N,N,5) -2.430932** -1.608175 stable
FSA (N,N,0) -0.810179 -1.609798 non- stable
DFSA (N,N,0) -5.872905*** -1.609571 stable
RI (C,N,0) 0.076889 -2.625121 non- stable
DRI (C,N,0) -4.073527*** -2.627420 stable
URB (C,N,1) 0.739071 -2.627420 non- stable
DURB (C,N,7) -2.632978 -2.650413 stable
Table 1 – ADF unit root test

Although the original sequence is not smooth, but after the first order difference
is stable, the same order single integer sequence is there may be a co integration
RISTI, N.º E5, 07/2016 381

relationship between them. Based on the VAR model, this paper carries out a Johansen
co integration test on the relationship between variables, and then determines the
structure of VAR model before this, see formula (1). Co integration test results are
shown in table 2.
Maximum
hypothesis eigenvalue LR 5% level
eigenvalue
R=0* 0.687860 58.61038(35.16623) 31.43624 29.79707
R≤1* 0.508586 27.17415(16.30449) 19.18262 15.49471
R≤2* 0.256199 7.991523(4.79491) 7.991523 3.841466
Table 2 – Co integration test
Co integration IG RFIR RFE

Not standardized 1.766774 21.13559 5.629345
Standardization 1.000000 11.96281 3.186228
(2.86665) (0.72417)
Table 3 – Co integration equation

Results from the table 2 can be explained, at 5% of the significant level, so the original
hypothesis has been rejected, and the adjusted LR values are larger than the critical
value, showing that there is more than two co integration relationship between the
variables. That there is a long-term equilibrium relationship between the income gap
between urban and rural areas and rural financial development, feature selection value
vector of the greatest as the rural financial development and urban-rural income gap
between the long-term cointegration equation, and the results are shown in table 3. The
co integration equation is:
IG = −11.96281 * RFIR − 3.186228 * RFE + 14.02024
The changes of rural financial development indicators of the size and the efficiency of the
rural financial development index of the income gap between urban and rural areas have
a negative effect, namely the promotion of rural financial development scale expansion
and the development of rural financial efficiency will reduce the income gap between
urban and rural areas. In the long run, the development of rural finance is conducive to
narrowing the income gap between urban and rural areas, and the integration process of
urban and rural areas is accelerated.
Although has been verified the cointegration relationship exists between the variables,
but it does not determine the causal direction between the variables. Therefore, this
paper will further study the causal relationship between the rural financial development
scale, efficiency and China’s income gap between urban and rural areas, using the
Granger causality test, the lag order number is 1, the Granger causality results are shown
in table 4.
382 RISTI, N.º E5, 07/2016

hypothesis F statistics Probability
RFIR is not the Grainger reason of IG 23.3362 6.E-05

IG is not the Grainger reason of RFIR 1.01493 0.3234
RFE is not the Grainger reason of IG 9.01422 0.0060
IG is not the Grainger reason of RFE 15.5344 0.0006
RFE is not the Grainger reason of RFIR 1.26585 0.2712
RFIR is not the Grainger reason of RFE 5.64553 0.0255
Table 4 – Grainger causality test

In the Granger causal relationship, we can get the following conclusions: the test result
is significant at the 1% level on the scale of rural financial development is the Grainger
reason of income gap between urban and rural areas of our country, Grainger reason of
income gap between urban and rural areas is not the scale of rural financial development;
rural financial development efficiency is the Grainger reason of income gap between
urban and rural areas, and in the 1% significance level, the income gap between urban
and rural in China is the Grainger reason of the efficiency of rural financial development,
the two have two-way causal relationship; Grainger causes the efficiency of rural financial
development is not the scale of financial development, while the reverse is significant
at the 5% level the scale of financial development is the Grainger reason of financial
development efficiency, increasing the scale of financial development the reason is to
enhance the efficiency of financial development.
4.4. Vector error correction model

Based on the VAR model, the vector error correction model is constructed to analyze the
short-term non equilibrium relationship between variables. In this paper to construct
the vector error correction model is to explain and analyze (fir, Fe) and development of
rural finance, fiscal expenditure for agriculture (FSA), fixed assets investment in rural
areas (RI) and urbanization (urb) of the income gap between urban and rural areas (Ig)
variable coefficient S.D T statistics

ECM(-1) 0.06932 0.03615 1.191748
D[IG(-1)] -0.14653 0.32248 -0.45440
D[RFIR(-1)] 0.42295 0.45006 0.93977
D[RFE(-1)] -0.17095 0.17290 -0.98874
FSA 0.00266 0.02412 0.11049
RI -0.00341 0.01481 -0.23028
URB -0.03397 0.02098 -1.61939
Adjusted R2 0.40293 S.E 0.10632
AIC -1.40353 SC -1.01958
Table 5 – VEC model estimation
RISTI, N.º E5, 07/2016 383

short-term fluctuation effect. Results are shown in table 5. From table 5 results show
partial coefficient is positive that the short-term fluctuations on the income gap between
urban and rural areas is a positive impact, namely short period of rural financial scale
expansion will widen the income gap between the urban and rural areas; rural financial
efficiency development RFE first-order difference of partial coefficient is negative, that
is conducive to narrowing the income gap between urban and rural areas to enhance the
short-term efficiency of rural finance.
4.5. Panel data model

The estimated results of the regression models in each region are shown in Table 6.
From table 6 shows the adjusted R2 values are above 95%, indicating that the data and
the regression equation of high goodness of fit, the value of F in the 1 per cent level
significantly, indicating that the explanatory variables on the explanatory variables
effect is generally significant, so this paper set model is more reasonable. According to
the Hausman test results, using fixed effect model equation. First, in the nationwide,
the effect of rural financial development scale and efficiency of the income gap between
urban and rural areas respectively in 10% and 1%significant level significantly and
present a negative effect, improve the rural financial development scale expansion and
the efficiency of the rural financial development, narrowing the income gap between
urban and rural areas.
Explanatory variable nation(1) East(2) Middle(3) west(4)

-0.019066* -0.026220* -0.011066 -0.037124*
rfir
(0.010702) (0.014462) (0.041291) (0.019848)
-0.128099*** -0.215943*** 0.476737*** -0.172511***
rfe
(0.039564) (0.051444) (0.159776) (0.058701)
0.494419 4.392752*** 0.690828 0.603672
urb
(0.365185) (0.541071) (0.867018) (0.513261)
-0.140028 -1.442697*** 0.968307* -0.126320
fsa
(0.285928) (0.299563) (0.504630) (0.416843)
-0.791643*** -1.430314*** 0.876015 -0.430886
ri
(0.264099) (0.285698) (1.076170) (0.531246)
3.061407*** 0.411221 2.177789*** 3.684554***
Constant
(0.175038) (0.330383) (0.481130) (0.206372)
Adjusted R 0.993316 0.987263 0.957768 0.987351
F-measure 652.221*** 280.044*** 74.705*** 281.999***
sample size 150 55 40 55
Table 6 – Panel model
5. References
In this paper, we describe the development of rural finance; by using time series theory
model and panel data model, we test the relationship between the variables. In the long
run, between the scale of the rural finance and rural financial efficiency and income
384 RISTI, N.º E5, 07/2016

gap between urban and rural areas exist stable relationship and were negatively to the
relationship that the rural financial development is conducive to narrowing the income
gap between urban and rural areas in our country. In theory, the development of rural
finance will promote the development of rural economy, so as to improve the income
of farmers, and ultimately reduce the income gap between urban and rural areas. The
empirical results and the theory are in full compliance with the. The scale of rural
financial development and the efficiency of rural financial development are the Grainger
reasons of the income gap between urban and rural areas, that is to say, both of them
are the reasons for the narrowing of the income gap between urban and rural areas. The
income gap between urban and rural areas and the scale of financial development is a
two-way Grainger causality, which shows that the income gap between urban and rural
areas is conducive to the improvement of the total scale of rural financial development.
In the short term, the impact of rural financial development scale expansion of the total
will widen the income gap between urban and rural areas, this may is the illusion of the
scale of the development of rural financial and rural deposit loans more than less and
disguised loans phenomenon make large capital not effectively put in the countryside.
The promotion of rural financial development efficiency will narrow the income gap
between urban and rural areas, and rural savings can be effectively transformed into
investment in rural areas, and create a favorable environment for economic growth. In
the nationwide, the effect of rural financial development scale and efficiency of the income
gap between urban and rural areas respectively in 10%, 1% significant level significantly,
and has a negative effect, as the rural financial development scale expansion and the
development of rural financial efficiency has improved, will narrow the urban-rural
income gap. Urbanization level and the impact of fiscal expenditure on rural income gap
is not obvious, there is a significant negative effect of investment in fixed assets in rural
areas, it can reduce the income gap between urban and rural areas.
Acknowledgments
This paper is supported by Social Science Fund Project of Heilongjiang province:
“Study on the evaluation and development of rural financial ecosystem in Heilongjiang
province”, (15JYD03).
References
Clark, G., Xu, L. (2006). Finance and income inequality: What do the data tell us?
Southern Economic Journal, 72 (3), 578–596.
Chen, P., Sun, Y. (2007). Rural financial reform and development under the change of
marginal constraints and cost structure. Management of the world, (3), 81–88.
Chen, W. (2009). Research on the “Kuznets effect” of financial development and urban
rural income distribution. Economic latitude and longitude, (1), 22–25.
(14), 1–15.
RISTI, N.º E5, 07/2016 385

Levine, R., Norman, L., (2000). Financial Intermediation and Growth: Casuality and
Causes. Journal of Monetary Economics, (46), 31–77.
Li, J. (2007). The loss of rural financial resources and the income gap between urban
and rural residents. Statistics and decision making, (10), 87–89.
Townsend, R., Kenichi, U. (2006). Financial deepening, inequality and growth: a model
based quantitative evaluation. Review of Economic Studies, (1), 251–293.
386 RISTI, N.º E5, 07/2016

Simulation of Influencing Factors in Internet-

Enabled Transformation Performance of Traditional
Enterprise Based on System Dynamics
Feng Ji, Ting Zhang, Fan Wu, YuPeng Mou
msjifeng@126.com
School of Management; China University of Mining & Technology; Xuzhou 221116; China
Pages: 387–399
Abstract: This paper regards traditional manufacturing enterprises as research

object, through analyzing the major Internet-enabled transformation elements, and
by means of questionnaires and expert interviews, uses rough set tool to determine
key influencing factors of Internet-enabled transformation implementation that
have a significant impact on the Internet-enabled transformation performance. In
order to more accurately grasp the dynamic relationship between Internet-enabled
transformation and influencing variable, this research uses system dynamics tools
to make feedback dynamic analysis of big data capacity, supply chain flexibility
and Internet-enabled transformation and establish a system dynamics model of
traditional manufacturing enterprise Internet-enabled transformation influencing
factors. And then a causality diagram and system flow chart of influencing variables,
transformation degree and transformation performance was constructed in this paper,
which constructs quantifies key variables by computer simulation, clears the structure
and dynamic relationship between influencing variables and simulates the influencing
degree of the variables on the Internet-enabled transformation performance.
Keywords: Internet-enabled transformation, system dynamics simulation,
influencing factors
1. Introduction
With the development of Internet technology and mobile Internet technology, Specifically
as the increasing demand for e-commerce, mobile commerce ( m-commerce),Internet-
enabled transformation has become the only way for the development of traditional
enterprises. In this situation, the development of Chinese enterprise is also affected
by the Internet industry, for example, technology research, production design,
manufacturing and even to marketing, financing, all these links change to centralization
and integration. For this, the boundaries between enterprises are slowly blurred; supply
chain and industry chain are developing towards the direction of resource sharing and
business collaboration integration and innovation. Therefore, enterprises must have
Internet thinking if they want to realize the Internet, only this can improve internal
control ,value chain and information technology capability of the whole life cycle, so as
to enhance sustainable competitive advantage across the enterprise( Hai-Jian LI,2014).
RISTI, N.º E5, 07/2016 387

Many scholars from home and abroad studied the Internet-enabled transformation
related issues of traditional enterprise with professional knowledge and methods
in different disciplines. Overall, these studies are more fragmented and mostly are
representational descriptive study; quantitative studies are relatively few, although
there are researches on the Internet-enabled transformation influencing factor,
mostly the relationship under static situations. Based on this, this paper conducted a
survey and evaluation about the effect of three levels the environment, business ability
characteristics and organizational features on Internet-enabled transformation, used the
rough set theory to reduce these factors and got 9 influencing factors which are relatively
important. However, the impact of these factors on Internet-enabled transformation will
change with the change of time and environment and there is some relationship between
these influencing variables. Seeing that the system dynamics is a computer simulation
studying the socio-economic system dynamic behavior, it does not rely on mathematical
logic deduction to get the answer, but establishes dynamic simulation model based on
actual observation information between the input variables and the output variables,
then uses the computer experiments to test whether the real behavior can reproduce
in a computer model system, in order to obtain a description of future behavior of
decision systems. For this purpose, the author uses the system dynamic tools to make
dynamic analysis of many influencing variables, and establishes the system dynamics
model of the Internet–enabled transformation influencing factors, which is to build on
causal diagram and system flow chart between influencing variables Internet–enabled
transformation degree and transformation performance etc. This way quantifies the key
variables, clarifies the structure and dynamic relationship between influencing variables,
simulates the influencing degree of various variables on Internet-enabled transformation
and explores the driving mechanism of the Internet-enabled transformation, from
which to find the regularity between the influencing factors and the Internet-enabled
transformation performance and provide the help of theory and methods for the
traditional manufacturing enterprises to imply the Internet-enabled transformation.
2. Internet-enabled Transformation Influencing Factor of

Traditional Manufacturing Enterprises
First of all, environmental uncertainty is the fundamental driving source of the Internet-
enabled transformation(Souder W E&Davies-Cooper R,1998). The Internet accelerates
the advent of the information economy and brings two significant changes in the
business world: the demand side, the customers are first high-energized by information,
leading to the transfer of power on all aspects of the value chain ,the consumers center in
economic activity in the first time; the supply side, the Internet has greatly enhanced the
mobility and penetration of information, reduced transaction costs, promoted the large-
scale social division of labor, collaboration, and rapidly gathered resources according
to market demand, Completing project tasks through online collaboration is very
popular mode (Jian Z Q, Wang C&Chen J H,2015). The current Internet is extending
from marketing and service sectors to R & D and manufacturing sectors. In fact, the
development of the Internet is from sales, marketing, services, which are more in contact
with consumers, highly openness link to the opening part of the high degree of contact,
low openness link penetration (Feng Ji, Fan Wu, Bangjun Wang,2016). The enterprises
388 RISTI, N.º E5, 07/2016

use some big data technology and relevant digital analysis tools to achieve a user-centric
mode conversion(Jangga R, Ali N M&Ismail M,2015).
Secondly, the data driving ability is the core driving element of the enterprise’s internet-
enabled transformation. To realize the core effect of data driving, the enterprises not
only need achieve data collection, but also need realize the flow of data across the whole
enterprise operation, and this flow is automated, without too much human intervention.
Only realizing data automation , the enterprise can truly realize the decision-making
data(James Manyika,2011). In addition, the flexible supply chain is the key to build
Internet-enabled transformation(Wai,2007). In the process of enterprise Internet, the
Internet of production processes not only need improve people’s intelligence level, but
also need pay attention to the interconnection of devices. To achieve intelligence, some
language systems and standard protocols should turn to the Internet, so the production
process is the most difficult part of the enterprise Internet, but also the current emphasis
and difficulty need to be broken. In the background of the Internet, enhancing the added
value of production processes is the key to realize the enterprise’s transformation and
development. Before the popularity of the Internet, the supply chain system is largely
a linear supply chain, reducing the cost oriented and collaboration range relatively is
limited. Because of the natural social collaboration property, the supply chain is facing
a huge challenge to the “Internet”(W Wang,2015). The biggest advantage of the Internet
is that it can support large-scale, community-based and real-time division of labor and
cooperation. It greatly improves the collaboration efficiency between the consumer
and enterprises, and the original pyramid or chain structure is being compressed on
a flat surface(Mandal S,2015). Ultimately, this makes the individual needs can directly
trigger the efficient collaborative value network consisted by various companies. When
the linear supply chain was transformed into information-driven collaborative value
network mesh, it will also mean the establishment of a new division of labor and a highly
collaborative supply chain system of socialization(Ji Feng,Zhang Ting&Wu Fan,2016).
Through expert interviews and field research, this research summarizes the main factors
affecting the traditional manufacturing enterprises’ Internet-enabled transformation
according to the actual situation of Internet-enabled transformation, designs the
questionnaire in view of the influencing factors of Internet-enabled transformation
and transformation performance involved. This research invites the respondents to
use Likert scale in a 5-point scoring for the Internet-enabled transformation rate. The
problem settings of this questionnaire are based on extensive literature research and field
research. This paper makes primaries through repeated research and discussion, and
then invites scholars and business executives to review, delete the fuzzy representation
and poor correlation topics, and modify certain questions that can be misleading or
ambiguous to make the final questionnaire items determined can accurately express
the desired purpose. To ensure the validity and accuracy of the survey data, we
reasonably arrange the sampling targets, cooperate with the investigation unit and the
questionnaires are hand-delivery etc. A total of 450 questionnaires were distributed and
408 questionnaires were recovered. Through judging the consistency and logic of the
questionnaire data, this research screened 400 valid questionnaires as the final data
sample. The questionnaire recovery rate was 90.67% and the effective rate was 88.9%.
RISTI, N.º E5, 07/2016 389

This paper uses SPSS to test the reliability and validity of the valid questionnaires
recovered. The Cronbach α reliability coefficient of this questionnaire is 0.928,
which shows good stability and reliability. Then this research respectively uses SPSS
software to test the structural validity of the Internet-enabled transformation factors
and transformation performance of this questionnaire, Kaiser Meyer-0lkin coefficients
were 0.873 and 0.908, the results show that this questionnaire has good reliability and
validity, and is basically consistent with the design requirements, the test results can
meet the statistical requirements and the level this research required. Although the
Internet-enabled transformation factors have very important impact on transformation
and its performance, there are many differences in the degree of the impact of such
factors on the transitional performance. Thus, on the premise of not changing the original
expression effect of the survey, this research uses the attribute reduction function of
rough set theory to delete redundant attributes. After reduction, this research discoveries
technology dynamic, demand dynamic, competitive hostility; resource integration
capability, depth analysis ability, real-time insight and scientific forecasting ability;
information flexibility, logistics flexibility, robust network and reconfigurable flexible
these 9 factors are relatively important for the traditional manufacturing enterprise to
achieve Internet-enabled transformation(Chen L F&Tsai C T,2016).
3. Simulation of Internet-Enabled Transformation

Performance Factors
3.1. Simulation analysis purposes
Many factors affect the traditional manufacturing enterprises’ Internet-enabled
transformation, in order to more accurately grasp the dynamic relationship between
the variables affecting the Internet-enabled transformation, this research uses
dynamic system method to construct the causality diagram and system flow chart of
the influencing variables, Internet-enabled degree and transformation performance
through identifying influencing variables the system involved; This simulation testing
can clarify the structure and dynamic relationships between variables, analyze the
motivation mechanism of the Internet-enabled transformation, find the regular patterns
between influencing factors and Internet-enabled transformation performance, and
provide theoretical and practical application support for the traditional manufacturing
enterprises’ Internet –enabled transformation.
3.2. System demarcation

First, we select the relevant state variables from the practical problems of Internet-
enabled transformation, classify the carrier by its determined state and determine
which state variables control the research variables, then we repeat this process until we
don’t have to pursue its arguments or its arguments can be ignored, so as to reach the
boundary systems. In view of this, this research makes the entire range of affecting the
traditional manufacturing enterprises’ Internet-enabled transformation as a boundary
of the system and the environment. The system contains all the key variables closely
related modeling purposes, we can select key subsystems and critical variables to
establish SD model. The research hopes to use the system dynamics simulation tool to
build the system dynamics flow chart of Internet –enabled transformation influencing
390 RISTI, N.º E5, 07/2016

factors and explore the law between influencing factors , the Internet-enabled level and
transformation performance. Through field research and questionnaire survey, the above
part determines the important factors affecting the Internet-enabled transformation.
In addition, the Internet-enabled level of traditional manufacturing enterprise also has
a significant impact on the transitional performance. Therefore, elements contained
by the system dynamics model are important factors affecting Internet –enabled
transformation, Internet-enabled degree of traditional manufacturing enterprises and
transitional performance.
3.3. System flow chart of influencing factors

SD causal diagram can help us understand the relationship between the various factors
and other factors and correctly establish variable equations and flow diagrams. There
are three major subsystems namely: environmental factors, big data capacity factors and
supply chain flexibility, affecting the overall level and they are inherently interrelated.
Through identifying influencing variables the system involved, from the perspective of
personalized marketing, flexible production, social supply chain, this paper makes in-
depth analysis of the relationship between the system elements. Building the system
dynamics flow chart of the Internet-enabled transformation factor is to establish a
causal link and feedback logic between many variables or elements in system, so that
these variables form a fully closed loop network making them connect and interact
with each other, and then we can establish a chain of causal feedback relationship and
mechanisms between the Internet-enabled transformation factors, Internet-enabled
degree and transitional performance. In the context of internet, data has become the
core driver of enterprise production and development. Only on the basis of data, can the
service links and flows be get through by the data, activating the data accumulated in the
production and operation of enterprises(Kastens K & Krumhansl R,2015). This is the key
value of data. Boosting by the development of enterprise and the Internet technology,
the enterprise has accumulated large amount of data, including structured data and semi
structured data. These data come from enterprise and outside the enterprise, including
competitor, user and market. How to use these data is the most important question,
and the developing and utilizing of data is the most crucial value. Understanding
supply chain flexibility should be based on environmental uncertainty, supply chain
flexibility brings not only scope economy but also the capability of dealing with market
demand uncertainty. Under the context of Internet, the core enterprise is facing with
many uncertainty factors, including market environment, consumers’ changing needs,
cooperation between supply chain partners, operation management inside enterprise,
stocks in every segments, supplier manufacturing quality, information delivering and
traffic condition( Fredriksson A, Wänström C.,2014). These uncertainty increases the
complexities of supply chain management.Based on the key factors analyzed above,
figure 1 presents the causality of the system.
For traditional enterprise, it’s hard for them to develop personalized marketing. However,
the capability of cloud computing and big data makes the personalized marketing
possible. The internet brings the enterprise more cost-effective and high efficiency
marketing method and reach numerous decentralized and personalized consumer
demands. These demands make up the large scale and personalized market. Consumers
can also express their demand and participate in different manufacturing sectors with
RISTI, N.º E5, 07/2016 391

Figure 1 – Proposed Framework Linking
Number Variable Meaning

Perceived industry technology evolution/technology developing
1 Technological dynamics
speed
2 Demand dynamics Uncertainty of perceived market demand or consumer demand
Lots of competitors, disadvantage supplies, rigorous controls, price
3 Hostile competition
war and lack of consumer loyalty
4 Resource integration Capability to obtain, integrate and update big data resource
5 Depth analysis Ability to deeply mining, analysis and predicting based on data
Real time observation Ability to real time observation and forecasting market changes
6
and scientific prediction based on data
7 Information flexibility Information sharing flexibility and information system flexibility
Ability to dealing with environment uncertainty, using proper
transportation and stock strategy to adjust service demand,
8 Logistics flexibility
stock space and logistics strategy, including stock flexibility and
transport flexibility
Robustness and The stability and solidity of current network structure when facing
9
reconstitution flexibility with changeable environment
10 Personalized Marketing Manufacture products according to consumer demands
11 Flexible manufacturing Manufacturing system swiftly response to market demands
12 Socialized supply chain Supply chain value network based on big data
The transformation
13 Consumer satisfaction, response rate, financial performance
performance
Table 1 – Meaning of Variables
392 RISTI, N.º E5, 07/2016

different forms. The evolution of manufacturing flexibility is the underlying revolution of

manufacturing industry since half the past century. The flexibility and rapid response is
the basic requirement of manufacturing industry. Without flexibility, the speed, quality
and cost cannot be balanced under the environment of internet(Zemirli N, Ali Alsenan
S,2015). Multispecies short run cannot be developed by the lack of economy. Because
of the social natural attributes of the supply chain, the current form of supply chain is
facing with the challenge of internet. Only by “dotcommed” and collaboration based
on the data shared on the internet can bring the high efficiency of collaboration and
decision (Jussila I, Tarkiainen A.,2015).
3.4. Simulation coefficients

We select the questionnaire data as the initial data of the transformation performance.
On the basis of stepwise discriminant analysis and Delphi expert judgment method, the
weight coefficients of every factor were determined. Using the product of initial data and
its weight as the model’s coefficients (Table 2).
Initial
Number Variable Weight Final Parameter
Parameter
1 Technological dynamics 4.21 4 16.84
2 Demand dynamics 4.46 4 17.84
3 Hostile competition 4.02 3 12.06
4 Resource integration 3.98 4 15.92
5 Depth analysis 3.76 4 15.04
Real time observation and scientific

6 3.67 5 18.35
prediction
7 Information flexibility 3.69 5 18.45
8 Logistics flexibility 4.01 4 16.04
Robustness and reconstitution

9 3.58 3 10.74
flexibility
10 Personalized Marketing 3.72 5 18.6
11 Flexible manufacturing 3.68 5 18.4
12 Socialized supply chain 3.85 4 15.4
13 The transformation performance 3.76 4 15.04
Table 2 – Parameters of influencing Factors

Note: The initial data is filtered from the 400 questionnaires.
System dynamic equations used in the paper are as follows:

•• The transformation performance=INTEG (performance changes,
initial data, a=3.92) (1)
RISTI, N.º E5, 07/2016 393

•• Performance Changes=b1* Personalized Marketing+b2* Flexible

manufacturing+ b3* Socialized supply chain (2)
•• Personalized Marketing=c1*Real time observation and scientific
prediction+c2*Depth analysis+c3* Flexible manufacturing+c4* Hostile
competition(3)
•• Flexible manufacturing=d1*Real time observation and scientific
prediction+d2*Depth analysis+d3*Information flexibility +d4*Logistics
flexibility+d5*Robustness and reconstitution flexibility+d6* Socialized
supply chain (4)
•• Socialized supply chain=e1*Resource integration+e2*Technological
dynamics+e3*Demand dynamics+e4* Hostile competition+e5*
Robustness and reconstitution flexibility (5)
•• Demand dynamics=f1* Technological dynamics (6)
•• Hostile competition=g1*Demand dynamics+g2*Flexible manufacturing (7)
•• Real time observation and scientific prediction=h1* Resource integration (8)
•• Depth analysis=i1* Resource integration+i2* Real time observation and
scientific prediction (9)
•• Information flexibility=j1*Resource integration+j2* Real time
observation and scientific prediction (10)
•• Logistics flexibility=k1*Information flexibility (11)
The initial data of performance change in equation 1 is 3.92. Coefficients in equation
2 to equation 11 equal to final parameter of independent variable multiply by the
weight of dependent variable and divided by final parameter of dependent variable. For
example, in equation 10, the weight of resource integration and real time observation
and scientific prediction to information flexibility (dependent variable) is m1, m2, and
j1=m1*15.92/34.27, j2=m2*18.35/34.27, the importance of independent variable to
dependent variable (such as m1, m2) are determined by expert scoring.
3.5. System simulation of influencing factors in Internet-based

transformation
According to on-site survey, the Internet-enabled transformation is a long-standing and
long-term process, the effect of transformation can be reflected after 3months adopting
big data, supply chain flexibility and socialized value chain. So the simulation period
used in this module is 6months (the values in x axis and y axis are derived from different
simulation points).
By comparing and analyzing simulated image when environment uncertainty level is
0.9, we find the curve of influencing factor of transformation performance is below
other curves. That means personalized marketing, flexible manufacturing, socialized
value chain, transformation performance are highly sensitive to influencing factor
of Internet-enabled transformation performance. In the figure of influencing factor
and personalized marketing, the opening degree of curve is high, meaning when the
environment uncertainty level is high, the changes of influencing factors have great
effect on personalized marketing.
394 RISTI, N.º E5, 07/2016

Figure 2 – Simulation of Internet-enabled Transformation Performance(Environment

Uncertainty Level=0. 9)
Note: Figure2 shows the variations of dependents (personalized marketing, flexible manufacturing, socialized
value chain, and transformation performance) during 6 months. The figures are calculated by the Vensim.
The influence factors of the dependent variables are arithmetic means of the simulation results. For example,
assuming the influencing factor (performance changes) of transformation performance (dependent variable)
is z, the independent variables are environment uncertainty, personalized marketing, flexible manufacturing,
and socialized value chain. The simulation results of each variables are x1, x2, x3, x4, so z= (x1+x2+x3+x4)/4.
The rest 4 influencing factors of dependent variables are the same computational methods.
Figure 3 shows the correlations between influencing factor of Internet-enabled

transformation performance and personalized marketing, flexible manufacturing,
socialized value chain, transformation performance.
By comparing and analyzing simulated image when environment uncertainty level
is 0.5, we find the changes of influencing factor of Internet-enabled transformation
performance has little effect transformation performance, but has great effect on
personalized marketing and flexible manufacturing. Notably, in the figure of influencing
factor and socialized value chain, with the increasing of Internet-enabled transformation
time, the sensitivity of influencing factor on socialized supply chain is increasing too.
By comparing figures when environment uncertainty level is 0.5 and 0.9, we find that
the change of influencing factor has little effect on transformation performance when
the environment uncertainty level is 0.5, but the change of influencing factor has great
effect on transformation performance when the environment uncertainty level is 0.9.
That means the transformation performance of traditional manufacturing industry can
be enhanced by environment judgment and forward-looking strategy planning.
RISTI, N.º E5, 07/2016 395

Figure 3 – Simulation of Internet-enabled transformation performance (Environment

Uncertainty Level=0. 5)
Note: Figure3 shows the variations of dependents (personalized marketing, flexible manufacturing, socialized
value chain, and transformation performance) during 6 months. The figures are calculated by the Vensim.
The influence factors of the dependent variables are arithmetic means of the simulation results. For example,
assuming the influencing factor (performance changes) of flexible manufacturing (dependent variable) is z, the
independent variables are information flexibility, logistics flexibility, robustness and reconstitution flexibility.
The simulation results of each variables are q1, q2, q3, so z= (q1+q2+q3)/4. The rest 4 influencing factors of
dependent variables are the same computational methods.
3.6. Model validation

1. Visual inspection:Visual inspection is mainly test whether the model is consistent
with the internal mechanism of the system, causality is reasonable and there is
correct definition for each variable. We specifically go to the Internet-enabled
transformation enterprise for line of research, during this period, we carry out
many communications with the senior management team about the system
dynamics model built in this paper. We repeatedly discuss the causal relation
and mechanism between each variable in the system flow chart. They make
objective judgments according to their practical implementation experience of
Internet-enabled transformation, and show that the causal diagram and system
flow chart this simulation model built is more reasonable. The state variables,
rate variables and the main feedback structure between variables basically can
describe the main features of the real system and reflect the actual operating
conditions(Luan K,2015).
396 RISTI, N.º E5, 07/2016

2. Operational inspection:Operational inspection is that using the checking and

tracking command of computer software to check the accuracy of the entire
system dynamics model (Wang B, Xia Z, Yu L, et al.,2015). Vensim software can
find errors in the model through functional test Check Model command, when
the model error occurs, the system will prompt the model is wrong during the
operation, you can not run the simulation, and the system prompts cause of the
error. Through repeated calibration and testing, this paper finally passed the
check model testing in the process of model simulation, which shows that the
system determines the boundary accurately, causal relationship of the model is
correct and reasonable and the information feedback network is in line with the
logic( Papadopoulos A V, Leva A.,2015).
3. Historical Inspection:Historical inspection is mainly to test whether the
simulation results are consistent with the past experience behavior of the actual
system. We conduct research on the enterprise implementing the Internet-
enabled transformation, interview with corporate customers, suppliers and
partners and get the validation. From a historical perspective, their analysis
shows that the system dynamics simulation results in this paper are basically
consistent with the actual experience of the Internet-enabled transformation.
4. Conclusion
As the Internet-enabled transformation process is a complex system, the effects of variables
on Internet-enabled transformation are affected by the changes of inside and outside
environment. Exploring the relationship between environment uncertainty, influencing
factor of Internet-enabled transformation and transformation performance has great
academic and managerial implications. We furthermore reveal the change of influencing
variables under dynamic situations by simulation analysing the reason, process and result of
the changes. This paper structures the dynamic model of influencing factor about Internet-
enabled transformation for traditional manufacturing industry. We analysed the relationship
between influencing factors and personalized marketing, flexible manufacturing, socialized
value chain and transformation performance. The result shows that, when the environment
uncertainty level is 0.9, personalized marketing, flexible manufacturing, socialized value
chain and transformation performance are highly sensitive to the changes of influencing
factors of Internet-enabled transformation, especially the personalized marketing. When
the environment uncertainty level is 0.5, the influencing factor of Internet-enabled
transformation has little effect on transformation performance, but has great effect on
personalized marketing and flexible manufacturing. We further find that the change of
influencing factor has more influence in transformation performance when the environment
uncertainty level is 0.9, this effect is not observed when the level is 0.5. That means different
environment uncertainty level has different influence on transformation performance.
Enterprises can boost transformation performance by improving their capacities to respond
to environment uncertainty, ability to forecast and adjust to changing circumstances. The
influencing factor has great influence on personalized marketing and flexible manufacturing
when the environment uncertainty level is 0.5 and 0.9. That means enterprises can enhance
their personalized marketing and flexible manufacturing level by improving their capabilities
of resource integration, depth analysis, real time observation and scientific prediction,
information flexibility, information flexibility, robustness and reconstitution flexibility.
RISTI, N.º E5, 07/2016 397

Acknowledgments
This paper is supported by Project of Philosophy and Social Science Research in Colleges
and Universities in Jiangsu Province (2016SJD630077); Jiangsu province education
science the 12th five-year plan project (B-b/2015/01/029); Graduate students teaching
reform project of China University of Mining & Technology (2015Y02, 2016Y03).
References
Chen, L., Tsai, C. T. (2016). Data mining framework based on rough set theory to
improve location selection decisions: A case study of a restaurant chain. Tourism
Feng, J., Fan, W., Bangjun, W. (2016). Research on the innovation ability evaluation of
traditional enterprise’s business model for internet transition with hesitant fuzzy
information. Journal of Intelligent and Fuzzy Systems, 31(1), 91–97.
Fredriksson, A., Wänström, C. (2014). Manufacturing and supply chain flexibility-
towards a tool to analyse production network coordination at operational level.
Strategic Outsourcing An International Journal, 7, 21–36.
Hai-Jian, L., Tian, Y., Wen-Jie, L. I. (2014). Mobile Internet Thinking and Traditional
Business Reengineering. China Industrial Economics, 8, 15–21.
Jian, Z., Wang, C. (2015). Strategic Orientation, Dynamic Capability and Technological
Innovation: Moderating Role of Environmental Uncertainty. R & D Management,
11, 232–246.
Jangga, R, Ali N M, Ismail M, et al. (2015). Effect of Environmental Uncertainty and
Supply Chain Flexibility Towards Supply Chain Innovation: An exploratory Study.
Procedia Economics & Finance, 31, 262–268.
Jussila, I., Tarkiainen, A. (2015). Individual Psychological Ownership: Concepts,
Evidence, and Implications for Research in Marketing. Journal of Marketing
Theory & Practice, 23(2), 121–139.
James, M., Michael C. (2011). Big data: The next frontier for innovation, competition,
and productivity. McKinley Global Institute, 11, 21–32.
Ji, F., Zhang, T. (2016). Review on the Relationship between Big data Capability, Supply
Chain Flexibility and Internet-based Industry Transformation Performance.
Academics, 2,68–78.
Kastens, K., Krumhansl, R., Baker, I. (2015). Thinking Big: Transitioning Your Students
from Working with Small, Student-Collected Data Sets toward Big Data. Science
Teacher, 7, 82–90.
Luan, K. (2015). Simulation of Freight Consolidation Strategy Based on System Dynamics.
ICLEM 2010@sLogistics For Sustained Economic Development: Infrastructure,
Information, Integration. ASCE, 3616–3622.
398 RISTI, N.º E5, 07/2016

Mandal, S .(2015). Towards an Empirical-Relational Model of Supply Chain Flexibility.

International Journal of Information Systems & Supply Chain Management, 8(3),
67–86.
Papadopoulos, A. V., Leva, A. (2015). A model partitioning method based on dynamic
decoupling for the efficient simulation of multibody systems. Multibody System
Dynamics, 34(2), 163–190.
Souder, W. E., Sherman, J., Davies-Cooper, R. (1998). Environmental uncertainty,
organizational integration, and new product development effectiveness: a test of
contingency theory. Journal of Product Innovation Management, 15(6), 520–533.
Wang, B., Xia, Z. (2015). Research of the Emergency information sharing in the Online
Social Networks Based on System Dynamics Simulation. Journal of Modern
Information, 10, 67–74.
Wai, P. (2007). Supply chain performance measurement system using DEA modeling.
Industrial Management & Data Systems, 107(3), 361–381.
Wang, W., He, R. (2015). The Data Service Mode of Remote Sensing Satellite Based on
the Internet Thinking. Science Technology & Industry, 7, 42–46.
Zemirli, N., Ali, A. (2015). PERSO-Retailer: Toward a Web Content Management System
Based on a Personalized Marketing Recommender System for Retailers. Cloud
Computing (ICCC), 2015 International Conference on IEEE, 56–60.
RISTI, N.º E5, 07/2016 399

Design of Injection Mold for Automobile Grille Based

on Computer-aided CAD Technology
Jinlian Wang
* wangjinlian83@126.com
Department of Mechanical and Electrical Engineering, Hangzhou Polytechnic, Hangzhou 311402,

Zhejiang, China
Pages: 400–412
Abstract: In this paper, the author analyzes the theory and method of injection
mold CAD/CAE technology, and completes a set of car grille injection mold design.
By using CAD/CAE technology in the mold design and manufacture, enterprise could
greatly improve the design level of plastic mold and product quality. The injection
molding process of plastic parts is simulated by MoldFlow software, and the author
makes evaluation of gating system and cooling system. Under the manual modeling
and Wizard Mold environment, we construct the solid model of each structural
component refers to the CAE analysis results. The result shows that application of
CAE/CAD technology to complete the design of injection mold for automobile grille
could avoid the common five kinds of defects of the grid plastic parts, and it will
improve the efficiency of the mold design.
Keywords: Computer-aided CAD, injection mold, structural parameters,
Simulation analysis
1. Introduction
With the development of productive forces, die industry has become one of the important
basic industries in the national economy. The traditional mold design and manufacturing
methods, mainly depends on the experience of designers and technicians. Mold
design is reasonable, products have no defects are only through the test mode to know
(Hinldsekhar, 1992; Bushko, 1996). Computer aided design, analysis and manufacturing
(CAD / CAE) technology in the computer the mold design for analysis and simulation
and prediction of potential defects in design help the designer to modify and optimize
design scheme, to improve the success rate of a test. CAD/CAE technology in the mold
design and manufacture of the application, greatly improve the level of plastic mold
design and product quality, and promote the development of China’s mold industry faster
(Gao, 2007; Delgado, 2015). CAD/CAE technology of injection mold is a mold design
and manufacturing personnel, in computer assisted, according to the injection molding
product geometry and non-geometry properties, to carry out injection mold design and
analysis of a kind of advanced technological means (Chen, 2013;Jiménez, 2015).CAD
system of plastic injection mold includes modeling of injection molding products, mold
design, part design, component assembly and structural checking; injection molding
400 RISTI, N.º E5, 07/2016

CAE is mainly the use of finite element technique to simulate the whole process of
injection molding, to die for liquidity analysis and cooling analysis, structure stress
analysis(Dai, 2014; Wen, 2014). The CAD/CAE system usually need plastic and mold
material performance parameter database, standard mold, injection molding machine
specification database, mold standard parts graphic library and database establishment.
Using CAD/CAE technology for injection mold design, compared with the traditional
way of work, has the advantages of the following five aspects.
•• Improve the mold quality: Through the use of computer systems to store the
technical information, scientific development of mold design. The use of CAD
technology, human computer interaction, can make people; machines play their
own characteristics, so that the mold design can be more accurate and rapid
completion. CAE technology can help designers to compare different design
programs and carry out, with the CAE technology, designers can ensure that the
design of the mold in the structure of the correctness and rationality. Without
technical personnel manual programming, CAE technology can be drawn
processing program, which helps to process the mold parts in the numerical
control processing equipment.
•• Improve the efficiency of mold production: The modularity of mold standard
parts in CAD system widely used, design program, in the database retrieval of
design parameters of convenience, computer automatic drawing of use factors
can greatly improve the design efficiency. With CAE technology to support the
mold design, designers can save a lot of waste in the time of repeated repair. In
addition, the high efficiency of CNC machine tools in the production process of
extensive use, can change the traditional mode of operation of the low efficiency
of the status quo.
•• Reduce the cost of production: The cost of labor is greatly reduced by the
use of automation. In the traditional way of working, it is inevitable to carry
on the repeated repair and mold test. In the process of injection molding,
computer simulation can be used to avoid this kind of situation. So as to reduce
the production cost, shorten the development cycle of new products, so that
products in the market more competitive
•• Promote innovation: CAD/CAE technology will be designed to liberate from
repeated calculations and drawings, to provide designers with the conditions
of creativity, so that it has more work time and energy to carry out creative
thinking and exploration.
•• Improve the management level of enterprises: CAD/CAE technology support
for the modernization of enterprise management system, make enterprise
management, scientific and quantitative analysis of enterprise products, reduce
the production of blindness, monitoring and enterprise balanced allocation
of resources, maximize the excavations of the mold production process in the
human, financial, and material function
2. Design of injection mold of automobile grille

The injection mold is composed of two parts, the moving mode and the fixed mode.
The moving mold is installed on the movable working table of the injection molding
RISTI, N.º E5, 07/2016 401

Design of Injection Mold for Automobile Grille Based on Computer-aided CAD Technology
machine, and the fixed mold is arranged on the fixed working table surface of the
injection molding machine. After the closing of the moving mold and the fixed mold,
the plastic is injected into the mold cavity through the pouring system. Usually after
the injection mold is separated, the product is stuck on the side of the moving die,
which is introduced from the mold by the ejection mechanism set in the moving mold.
Injection mold basically can be divided into six parts: part of the molding, casting
system, exhaust system, temperature control system, remolding mechanism and
mold. Forming part of the structure of the cavity, and determines the plastic parts of
the shape, size and precision; pouring system as channel, by the nozzle of the injection
molding machine out of molten plastic drainage to the cavity; the exhaust system is
exclude the gas in the cavity in the mold filling; temperature control system controls
the temperature of the mold, plastic melting can in the mold cavity is filled quickly
and reliably stereotypes; demoulding mechanism to finalize the design of plastic parts
from the mold off points and remove parts; parts of the mold the mold effectively
connected together, and contact with the injection molding machine together, the
correct orientation and positioning.
Figure 1 – The basic structure of injection mold
The four main stages of the injection molding process are plastic, filling, packing and
cooling. In these stages, the key process parameters are not the same.
1. Plastic: plastic is a process of heating and being in a state of plastic flow. The
melt injection mold cavity should be heated to the required temperature, the
temperature requirement is uniform. In quantity, it is required to reach a
sufficient value within the specified time. In order to ensure that the molding
process is carried out smoothly, the melt does not occur or rarely occurs.
402 RISTI, N.º E5, 07/2016

2. Fill: This stage started from the screw or plunger is moved forward, to model
cavity filled with plastic melt check. The main process parameters of this process
are injection pressure, melt temperature and filling time. At the beginning of
mold filling, the pressure in the cavity is zero, and the pressure value is filled
in the cavity. The time required for filling the mold will affect the pressure
required for mold filling. The longer the time, the greater the viscosity of the
plastic injected into the mold prior to the cooling of the relationship, in order to
overcome the resistance, must be applied to the subsequent entry of the plastic
higher pressure. Mold filling time is short; the mold filling pressure is small.
3. Holding pressure: This stage is filled with the melt from the mold cavity, to the
screw or plunger back up. The important parameters of this stage are holding time
and pressure. At this stage, cooling will make the plastic melt contraction of the
phenomenon, however, due to the screw and plunger sustained pressure, melt
to is injected into the mold cavity, due to shrinkage voids can be compensated.
In addition, the plastic is also in a state of flow, at this time, the temperature
will continue to reduce the directional molecules are prone to the phenomenon
of freezing. So this stage is the main stage of the formation of large molecules.
4. Cooling: The important parameter of this stage is the cooling time. In the
cooling stage, there is no plastic flow out from the gate, but there may be a
small amount of plastic in the mold flow, and thus still may produce a certain
amount of molecular orientation. In a stage, the plastic pressure, temperature
and volume of the mold will change, so until the release, the pressure inside the
mold is not necessarily equal to the external pressure, the pressure difference
will produce residual pressure. In order to make the mold release smoothly and
obtain satisfactory products, the residual pressure must be close to zero.
The grille is a front ventilation grille of a micro car, as shown in Figure 2.2. Materials
for polypropylene (PP), product size of 690 x 22 x 150 in accordance with the traditional
method, the idea of plastic products completed, the need to create a solid model to evaluate
its appearance, the performance of its measurement. In addition, if only by intuition and
experience in the design of the mold, it is often need to repeatedly test mode and the
amendment, in order to make the mold to produce qualified plastic products. Computer
aided design and analysis can fundamentally change the mold of the traditional mode
of production and process, using the geometric modelling technology, product shape
and every part of the mould can vividly displayed on the computer screen; using finite
element analysis technology, products. The mechanics and mechanics performance can
be predicted, the injection molding process simulated.
Figure 2 – Vehicle grid solid model
RISTI, N.º E5, 07/2016 403

The cars on the grid of injection mold design using computer technology, the computer
aided design and analysis, the CAD of the injection mould for the automobile grille is mainly
completed grid size checking of the parts of the 3D modeling, gate injection mold structure
of 3D modeling, grid injection mold 3D digital to analog assembly and gate injection mold
structure are important parts in plastic. Geometric modeling (also known as 3D modeling)
is the core of the CAD system, which is the shape of the object and its attributes (such as
color, texture, etc.) stored in the computer, the formation of the three-dimensional model
of the object. Cars on the grid of the injection mould CAE is mainly completed cars on
the grid structure rationality test, determine the injection mold gating system and cooling
system structure of the program, determine the reasonable parameters of injection molding
process. The injection mould CAE system is the actual structure of discrete processing.
3. Computer simulation and automobile grille mold (CAE)

3.1. Moldflow injection mold design
In this paper, the Moldflow injection CAE software, in the injection mold design, the role
is mainly reflected in the following three aspects:
1. Optimization of plastic products: The location and quantity of the gate, the
thickness of the plastic part and the design of the runner system will directly affect
the quality of the product. In the traditional way of working, the engineering
technical personnel rely on experience to design is usually inefficient. By using
Moldflow software, we can optimize the structure of products, design more
reasonable products, and reduce costs, save production time.
2. Optimize the mold structure: In the process of its die design, the repeated test
mode, the repair module often cannot be avoided. Using Moldflow software
design can compare various design scheme, the gate number and location of the
optimization, the reasonable design of the runner system and cooling system,
the optimization of gate size and cavity size, the cooling system size and the size
of the flow passage. Compared with the traditional working mode, the number
of trimming die can be greatly reduced, and the quality of die can be improved.
3. Optimization of injection molding process parameters: Limited experience,
engineering and technical personnel are often difficult to accurately select the
reasonable processing parameters of products, the use of appropriate plastic to
determine the best processing program. Moldflow software, can help engineers
to determine the best pressure holding pressure, injection pressure, melt
temperature, injection time and other processing parameters, is conducive to
the processing of the best products.
3.2. Modeling simulation analysis of grid plastic parts

For any injection molding, the most important is to control the flow of plastic in the
mold. Many of the disadvantages of the product, such as cavitation, weld line, product
deformation, and plastic in the mold flow.. The grille plastic molding simulation analysis,
using MF/Flow flow analysis module of analysis of plastic in the mold flow state, filling,
weldline and bubble position, the shrinkage and warpage defect prediction; determine
appropriate process parameters of injection and the best solution casting system. In
the cooling system of a given design parameters, the injection mold cooling analysis
404 RISTI, N.º E5, 07/2016

software can predict the cooling medium flow state, die wall temperature distribution
and the cooling time, etc., design personnel assessment Gu design scheme, provides
an advanced and practical tool in optimizing the design of the cooling system. In the
simulation analysis of the grid plastic parts, the MF/Cool cooling analysis module was
used to analyze the cooling condition of the plastic parts, and to predict the final surface
quality, the residual stress and the degree of crystallinity.
Finite element partitioning is the division of the mesh of the solid model, which is a
small piece of a cell or a mesh. Grid division is very important in the whole process of
CAE, which will directly affect the accuracy and economy of CAE. The grid 3D number
of modules into MPI, Moldflow will be automatically divided into its finite element. The
results of finite element analysis are shown in Figure 3.
Figure 3 – Finite element method

For the complex shape, high precision and long service life of large and medium-sized
plastic mold, mold material is mainly used in the pre hardened plastic die steel. This
type of steel in the steel mill after fully forged into the module, pre heat treatment to the
requirements of the hardness, the general conditioning to 30~5HRC after use, P20, 718 is
the most widely used in the international pre hardened plastic mold steel. The grid plastic
part selects the P20 mold steel, the P20 has the good machinability and the mirror surface
grinding performance. The properties are: the density is 7.8g/cm3, the elastic modulus
is 2.0*105Mpa. After nitriding treatment, can get high hardness of the surface layer, the
surface hardness of 650~700HV (57~60HRC), die life can reach more than 100 times.
3.3. Pouring Scheme

Because grid plastic pieces of a larger, PP material liquid medium, in order to avoid plastic in
the pouring channel surface condensation, reduce loss of injection pressure, reduce the loss
of raw materials, reduce the product cost, using hot runner form runner; flow channel using
a circular runner, can reduce the pressure loss and heat transfer loss; gate are symmetrically
arranged at two sides of plastic parts, avoid short shot, can respectively flow from left to
right, balance of fill to melt into the cavity; gate using submarine gate, from plastic pieces of
the inner surface of the feed gate can again when the mould is opened automatically cut off,
and the gate after the removal of residual traces of small, good surface quality.
RISTI, N.º E5, 07/2016 405

Three casting schemes are designed in this paper.

1. Plan one: gate location is selected in the plastic part of the long direction of
the outside, the two side gate, gate diameter 4.5mm, flow channel diameter of
12mm;
2. Plan Two: the gate location is selected in the plastic part width direction outside,
the two side gate, the gate diameter 4.5mm, the channel diameter is 12mm;
3. Plan Three: the gate location is selected in the plastic part width direction
outside, is located two side gate, the gate diameter 6.5mm and the channel
diameter is 12mm.
Figure 4 – Pouring Scheme plan one
Figure 5 – Pouring Scheme plan two
MPI / flow through simulation of melt flow behavior in mold, plastic melt in the mold
flow analysis can predict and display the melt front propulsion, filling pressure and
temperature variations in the process, cavitation and weld line position, in order to
help design personnel before the test mode of possible defects prediction, find out the
causes of defects and improved in order to determine the appropriate gating system.
The filling time shows that the flow front position of every certain interval when filling
in the mold cavity shows the time required for the molten body to fill the whole cavity.
At the beginning of the fill, the color is dark blue, and the place is red. A balanced
fill, the color is evenly spaced. The analysis results of the three schemes are shown in
figure 6-8.
406 RISTI, N.º E5, 07/2016

Figure 6 – Filling time (scheme 1)
As can be seen from the figure, the three programs can be filled with good results. Three
scheme color is evenly spaced, there is no broken and hysteresis phenomenon, filling
is balanced. The filling time of the scheme is 1.306s, the filling time of scheme two is
1.396s, and the filling time of scheme two is 1.394s. Filling time difference is small.
RISTI, N.º E5, 07/2016 407

4. Computer aided design of injection mould of car grille

4.1. CAD modeling function
Wizard UG is a professional injection mold design module of Mold software, which
provides a convenient and practical tool for injection mold design. Using these
tools, you can design a set of complete mold structure, including type core, cavity,
formwork, inserts, a slide block, a gating system and cooling system. Maximum use is
the integration of a set of complete mold standard parts library, standard parts library
integrated multiple famous mold parts production factory standard parts, as long as
the corresponding parameter settings can call components and automatic assembly to a
suitable location, the trouble of replacing parts drawing and re assembly. Injection mold
of automobile grille adopts two plate mould design, mould is installed in the horizontal
injection molding machine, mold is fixed in the plastic injection molding machine fixed
template, mold is fixed in the plastic injection molding machine moving template, by
moving the moving platen. Complete the die mold, injection molding machine lock
mould mechanism to complete the mold locking. Then molten plastic from the injection
molding machine nozzle, hot runner system is pressed into the mold cavity, after a certain
time, molten plastic in the mold hardening shaping, mobile template mobile open mold,
injection molding machine of the top rod demoulding mechanism in promoting ejector
grid plastic parts. The injection mold of the automobile grille is shown in figure 9.
Figure 9 – Automobile grid injection mold

Parting is the design of the injection mould for the part of the most important, it is parting
surface will contain the volume of the mold cavity block separated into a dynamic model
of the core and cavity model, the parting surface is dynamic model of the core and the
contact surface of the cavity model. Because of the larger size of the grid, the mold is
decided to adopt a mold cavity. In the working environment of the mold wizard, use related
commands, UG system will automatically of plastic parts size and injection molding
material thermal expansion shrinkage parameters calculation, get the parting surface.
408 RISTI, N.º E5, 07/2016

Figure 10 – Parting surface
4.2. Structure of forming part

The parts which are directly related to the plastic to form a plastic part are called as
forming parts, in which the forming part of the plastic part is formed into a mold cavity,
and the forming part of the internal shape of the plastic part is made into a core. Due
to the larger size of the grid plastic parts, to ensure the rigidity of the mold, the overall
type of core and cavity. After the parting surface is completed, the mould core and the
mould cavity can be created by parting tool. UG will be based on the structure of the sub
type of the size of the automatic processing, generating cavity and core, and then in the
modeling environment, through the manual modeling method, complete the grid cavity
and core. The cavity and core of the grid are shown in figure 11 and figure 12.
Figure 11 – Cavity structure
Figure 12 – Core structure
RISTI, N.º E5, 07/2016 409

4.3. Gating system design

According to the structural requirements of plastic parts, the design of the gate, Latent
gate runner in the parting surface, and the gate itself in concealment of a mold, the
plastic melt through the cavity side inclined into the cavity, and plastic parts and no
damage to the surface, the gate mark will not affect the plastic parts of the surface quality
and beautiful. In the Wizard environment, using the relevant commands set the gate.
The gating system of grid injection mold is shown in figure 13.
Figure 13 – gating system
4.4. Ejection mechanism and cooling system

Each cycle of injection molding, products must be accurate and reliable from the cavity
or core. This is the demoulding mechanism of device to complete the process of extrusion
products. The design of ejection mechanism should follow the following principles: as
far as possible to make the parts in the moving die side, so that the. Clamping force
driving demoulding device; preventing product deterioration and damage; strive
for good appearance of products, launched as the location selected internal WIP or
appearance has little influence on the position, mechanism is reasonable and reliable,
and easy fabrication. In wizard environment, select “standard mould base management”
to choose straight push rod and set size, and then pruning the all of the push rod, the
system will automatically push rod trimmed to parting surface, putting complete design.
The ejection mechanism of the grid injection mold is shown in figure 14
Figure 14 – Ejection mechanism
410 RISTI, N.º E5, 07/2016

According to the analysis results of CAE, the cooling system is designed in four, and the
cooling system is designed. Grille injection mold cooling system diagram, such as figure 15.
Figure 15 – Cooling system

In the gating system and cooling system design process by before the stage of CAE provides
feasibility project of the structure of the mould, rational use of UG of other functional
modules, combined with mold Wizard provides mold standard parts library technology,
and effectively improve the development efficiency of the die design. Production practice
has proved that the use of CAE/CAD technology to complete the injection mold design of
the automobile grille, can effectively avoid the common five kinds of defects of the grid
plastic parts, the production of qualified Auto Grid plastic parts. In the injection process
of the injection mold, from the mold to the injection pressure, the cavity will be high
impact force, to improve the quality of mold design and mold to ensure the service life
of the application, using the reliability design theory, finite element analysis and fatigue
analysis technique of mold cavity design to continue in-depth study.
5. Conclusion
In this paper, the theory and method of injection mold CAD/CAE technology research,
combined with the injection mold design theory, completed a set of car grille injection
mold design. The method and process of the application of CAD/CAE technology in the
injection mold design are explored, which provides a reasonable reference for the design
of other shell plastic parts. The design principle of injection mold CAD/CAE technology,
the design principle of injection mold and the characteristics of five common defects in
plastic injection molding process are studied. The injection molding process of plastic
parts is simulated by using MoldFlow software. The flow and cooling analysis, five kinds
of defects which may appear in the grid plastic parts are predicted, the gating system
and cooling system scheme are evaluated, and the reasonable die casting system and
the cooling system structure are determined. Using UG NX software for injection mold
design, under the environment of manual modeling and mold wizard, reference of CAE
analysis results to construct grid injection mold of the components of the solid model,
accurate, rapid completed cars on the grid injection mold design. It has been proved that
the cars on the grid injection mold design is done using CAE/CAD technology, which
can effectively avoid grille plastic parts of the five kinds of common defects, effectively
RISTI, N.º E5, 07/2016 411

improve the efficiency of mold design and development, to produce plastic parts of the
qualified cars on the grid.
References
Bushko, W. (1996). Solidification of thermoviscoelastic melts. PartII: Effect of processing
conditions on shrinkage and residual stress. Polymer Engineering and Science.
(36), 352–364.
Chen, Y. (2013). SUN work station on the mold CAD/CAE. Journal of Harbin University
of Science and Technology, (03), 8–10.
Dai, C. (2014). CAD/CAM latest technology analysis. Computer aided design and
manufacturing, (11), 7–9.
Gao, Z., Cao, J. (2007). Analysis and application of plastic flow based on MoldFlow.
Mould manufacturing, 2007, (02), 10–11.
Hinldsekhar, K., Lottey, W. (1992). CAD of mold cooling in injeetion mold using a
there-dimensional numerieal simulation. Jounral of Manufaeturing Soienee and
Engineering, (144), 213–221.
Wen, J., Ma, X. (2014). Application of CAD/CAE software in injection mold technology.
Mold technology, (5), 17–27.
412 RISTI, N.º E5, 07/2016

Research on Rolling Bearing Fault Diagnosis Based

on Support Vector Machine
Shusen Li 1,*, Qingchun Zhang 1, Xin Shang 2, Bin-bin zhang2
* 99483865@qq.com
1
Northeast Forestry University, Harbin 150040, China
2
Harbin Normal University, Harbin 150025, China
Pages: 413–424
Abstract: In this paper, the authors analyze rolling bearing failure modes based on
wavelet denoising and SVM technology. By using a rolling bearing fault intelligent
diagnosis method based on support vector machine, simulation experiments shows
that due to the bearing signal is mixed with a lot of noise signal, using wavelet
denoising technology can effectively remove the original signal in the noise signal,
and effectively reduce the error. In view of the limited number of fault samples, we
use wavelet threshold denoising and support vector machine combination method
to diagnose the fault of rolling bearing. The experimental results show that the
method can quickly and accurately identify rolling bearing fault type.
Keywords: Support vector machine, rolling bearing, wavelet analysis,
fault diagnosis
1. Introduction
The rolling bearing is one of the most widely used and easily damaged parts in the
machine. According to statistics in the use of rolling bearing in mechanical equipment,
about 30% of the mechanical fault is related to the bearing fault and bearing fault not
easy to be found, this will of machinery and equipment time damage, rolling bearing
condition and directly affects the running state of the whole machinery and equipment
and the entire production line, thereby affecting the efficiency of the plant operations and
profit. Therefore, it is of practical significance to carry on working condition monitoring
and fault diagnosis of rolling bearings (Abhijit, 2005; Delgado, 2015).
Rolling bearing is one of the most important parts of mechanical equipment, and its
failure rate is very high. The working condition of the bearing affects the operation of
the whole equipment and even the entire production line (Farshid, 2007; Rai, 2007).
Therefore, it is very necessary and important to study the fault diagnosis of rolling bearing.
In this paper, the rolling bearing is studied, the basic structure and the failure mode of
the rolling bearing are analyzed, and the fault mechanism and vibration characteristics
of the bearing are studied (Yunyun, 2011). According to the characteristics of the rolling
bearing fault signal, the acceleration sensor to detect the vibration signal, of rolling
bearings of four kinds of working state: normal operation, inner ring fault, outer fault
RISTI, N.º E5, 07/2016 413

Research on Rolling Bearing Fault Diagnosis Based on Support Vector Machine
and rolling element fault were vibration signal acquisition. Because the traditional fault
diagnosis needs a large number of data samples, but the real test data samples are not easy
to get, which brings great difficulty to the fault diagnosis. So in this paper, we introduce
the support vector machine to fault diagnosis, and provide a new research method for
fault intelligent diagnosis (Sugumaran, 2007; Xiaodong, 2011). In this paper, we use
a combination of theoretical research and computer simulation research methods,
first of all with the vibration signal of the sensor bearing. Then the threshold method
of wavelet transform wavelet de-noising of rolling bearing vibration signal processing,
signal interference signal is removed. Then use the wavelet packet technology was used
to extract the noise signal frequency band energy. At last, the wavelet packet analysis get
the frequency band energy feature sets as input vector for support vector machine using
support vector machine intelligent classification judges the working state of the bearing.
2. Vibration mechanism of rolling bearing

The basic structure of the rolling bearing is shown in figure 1. It is composed of four
parts: inner ring, outer ring, cage and rolling element. The general inner ring is fixed on
the shaft neck, and the inner ring and the shaft are rotated together, and the outer ring
is used for assembling the bearing seat hole.
Figure 1 – The basic structure of rolling bearing

1. Pitting corrosion: In the normal operation of the bearing, rolling by the load
size different, resulting in rolling contact surfaces of the body and the inner and
outer contact force changes, when the contact stress is beyond the limit value,
the surface fatigue crack is generated and continuous expansion to the bearing
surface formation of pitting corrosion. Pitting can cause noise, shock and
vibration of the bearing to reduce the rotation precision of the rolling bearings,
and the pitting corrosion is the main failure mode of the rolling bearing.
2. Erosion: It is a kind of plastic deformation, which is caused by the heavy load
during the operation of the impact load or the stop.
3. Friction and corrosion of bearing: In fact, it is the corrosive wear of the
bearing. It is caused by the vibration of the transport link before the bearing
assembly and application installation. Vibration load caused by vibration load
can also cause this situation.
4. Corrosion caused by the current through the bearing: When the motor
414 RISTI, N.º E5, 07/2016

is used, the bearing is not insulated, and the current will pass through the
bearing and form a series of micro pits on the surface of the roller.
5. Indentation caused by hard particles: Hard particles into the lubricating
oil and bearings can also lead to the destruction of the surface of the rolling
contact surface.
Bearing at a certain load and speed, the bearing, the bearing seat of the vibration
system will produce vibration; show the mechanism of its vibration. The vibration,
including internal factors and external factors, internal is determined by bearing
structure characteristics, the rotation failure and other factors; external factors for
bearings where the drive shaft and the shaft on the other parts of the integrated effect.
Rolling bearing signal acquisition is the vibration signal obtained by installing sensors
in the bearing seat, which is collected by the internal and external factors together with
the integrated vibration signal of the bearing system. Therefore, the vibration signal
generated by the rolling bearing fault is separated from the comprehensive vibration,
which is the key part of the bearing fault diagnosis, which has a great influence on the
accuracy of the diagnosis. The natural frequency is only related to the characteristics
of the parts, and has nothing to do with the rotational speed of the rolling bearing. We
calculated the natural frequency when considering the internal factors of its material,
size, structure etc.. The natural frequency is much higher than the frequency of the
fault feature. The calculation formulas of the natural frequency of the rolling bearing
components are as follows:
f n = 9.4 × 105
(
hn n 2 − 1 ) (1)
b2 n 2 + 1
In the formula, n represents the order of the natural frequency, where n starts from 2:
2,3,4,... H, said the ring thickness in millimeters, B, said the ring width in millimeters.
From the above formula can be known, the bearing’s width is smaller, the thickness is
bigger, the inherent frequency of the inner and outer ring is bigger. Natural frequency
of rolling elements:
0.424 E
fb = (2)
γ 2ρ
E means the density of the material elastic modulus, ρ means rolling radius. γ means
the larger the elastic modulus, the greater the material density and the radius, the
greater the inherent frequency of the rolling element. When the inner ring, outer ring
and the rolling body wear fault, with rotation of the bearing, its surface will produce the
composition of the low frequency vibration, which is generated by surface of repeated
impact, the periodic pulse vibration time is short, steep shape, called for “by vibration,
the vibration frequency, also known for rolling bearing fault characteristic frequency can
obtained through the bearing size and speed. This type of fault signal is also the main
object of this thesis. After the extraction of the actual bearing signals, the classification
and recognition ability of the support vector machine is to determine which kind of fault.
According to four different damage locations of rolling bearings: rolling bearing inner
RISTI, N.º E5, 07/2016 415

and outer ring, rolling body and cage, the formula for calculating the fault characteristic
frequency is given in table 1.
Damage location characteristic frequency
z  d 
inner ring fi = f a  1 + cosα 
2  D 
z  d 
outer ring fo = fa 1 − cosα 
2  D 
D  d2 
Rolling body fb = f a  1 − 2 cos2 α 
2d  D 
1  d 
fc = f a 1 + cosα  z (touch the inner ring)
2  D 
Holder
1  d 
fc = f a 1 − cosα  z (touch the outer ring）
2  D 
Table 1 – Calculation formula of Characteristic frequency for rolling bearing

In table 1, fa is the turning frequency, α for the contact angle, d for the rolling body
diameter, z for the number of rolling elements, D for the bearing diameter. But in practical
applications, the fault characteristic frequency and the theoretical value obtained after
the signal analysis are not exactly the same.
3. Noise reduction and wavelet packet energy extraction of rolling

bearing
3.1. Wavelet analysis
If Ψ(t)∈L2(R) to meet the permit conditions:

∞ ψ (ω )
Cψ = ∫ dω < ∞ (3)
−∞ ω
Then Ψ(t) called base wavelet or admissible wavelet. Based on the basis of wavelet
function, we derive the formula of wavelet function, which can be expressed as:
 t −b 
ψ a ,b ( t ) = α
−1 2
ψ  (4)
 a 
Through this function to decompose the signal processing, the decomposition process is
the continuous wavelet transform.
416 RISTI, N.º E5, 07/2016

(W f ) (t ) = ∫ f (t )ψ (t )dt (5)
∞

ψ q ,b
−∞
We again without loss of information under the premise, the parameters a and b or at
the same time discretization, this conversion for the discrete wavelet transform, discrete
wavelet function can be expressed as:
(
ψ j ,k ( t ) = a01 j 2ψ a0− j t − kb (6) )
The coefficients of discrete wavelet transform can be expressed as:
∞
W j ,k ( t ) = ∫ f ( t )ψ j ,k ( t )dt (7)
−∞
Signal reconstruction formula:
∞ ∞
f (t ) = C ∑ ∑W
−∞ −∞
ψ j ,k ( t ) (8)
f ,k
The signal was decomposed using a wavelet. In this paper, the DB4 wavelet three layer
wavelet decomposition of noise signal with, obtained after the decomposition of the first
layer to the third layer detail coefficient selected an appropriate threshold and threshold
processing is divided into hard threshold and soft threshold method, the hard threshold
method is the absolute value is less than the threshold of detail coefficients become zero,
but in the process will appear between the breakpoint. Soft threshold method for the
improvement of the shortcomings of the hard threshold method, the boundary is not set
to 0 points, so that the signal can be reconstructed without a break point, and smoother.
Hard threshold formula:
 w ( i, j ) ≥ T

W ( i, j ) =  (9)
 w ( i , j ) ≤ T
Soft threshold formula:
sgn w ( i , j ) w ( i , j ) − T

W ( i, j ) =  (10)
 w ( i, j ) ≤ T
Wavelet denoising is an important part of how to select the threshold, if the
threshold is too small, the effect of noise reduction will be affected, to reduce
the follow-up signal fault recognition accuracy rate. If the threshold selection is
too large, although some of doping doping the noise signal can be removed, but
effective vibration signal will be mistaken for a signal to noise and be disposed of, so
that the de noised signal distortion, cannot really shows the signal characteristics,
misleading fault type diagnosis.
RISTI, N.º E5, 07/2016 417

3.2. Experimental simulation of wavelet noise reduction

The quality of internal control not only has a direct impact on investment efficiency,
but also has an indirect impact. The indirect impact affects that internal control
quality through the financing constraints, agency costs and the quality of accounting
information produce a certain impact and then affect the efficiency of investment.
Wavelet denoising simulation using Daubechies wavelet DB4 wavelet family, it is
better to extend the wavelet, suitable for wavelet denoising. Due to the threshold
selection and quantification of has a direct influence on the effect of wavelet
denoising, the threshold is too small will make noise cannot be removed and lost the
purpose of noise reduction; if the threshold is too large, may will make the loss of
the components of the original signal, so we choose the default threshold de-noising.
Matlab simulation using wdencmp global threshold command, the decomposition
coefficients by parameters THR, SORH specified threshold by means of wavelet
function name with a specified wavelet. Figure 2-5 in order to normal bearing, inner
ring fault, and outer ring fault, the original fault of the rolling body signal contrast
after signal and noise reduction.
Figure 2 – Noise reduction contrast chart of normal bearing signal
Figure 3 – Noise reduction contrast chart of inner fault of bearing
418 RISTI, N.º E5, 07/2016

Figure 4 – Noise reduction contrast chart of outer inner fault of bearing
Figure 5 – Noise reduction contrast chart of rolling body fault of bearing
4. Fault recognition based on support vector machine

4.1. Support vector machine
SVM is a method based on statistical learning theory and structural risk minimization
theory, which provides a new way to solve the problem of small sample data classification
and nonlinear problem. SVM is peacekeeping linear rise of, through nonlinear mapping;
the sample space is mapped to a higher dimensional space within, the samples in high
dimensional space correctly classified, so that you can through linear learning machine
Figure 6 – Linear inseparable by increasing dimension into linear separable
RISTI, N.º E5, 07/2016 419

method to solve the problem of nonlinear classification in the sample space. Ascending
dimension is mapping the samples to higher dimensional space, which can be linearized
by linear hyper plane in high dimensional space, which can solve the problem that the
sample is not easy to be classified in the low dimension space. For example, figure 6
shows a typical example: randomly distributed points in the two-dimensional space to
linear programming, but by mapping to 3D space to achieve linear programming.
Here we only consider a loss function of SVM; the original problem is converted as:
m
1
φ ( w,ξ ) = ∑
2
w + C ξ i (11)
2 i =1
The problem is to minimize the constraints, That is:

2
w m
min
2
+C ∑ξ i =1
i
(12)

yi (( w ⋅ x ) + b ) − 1 ≥ 0 (13)
Then carries on the derivation of the dual problem, need to introduce the original
problem of Lagrange’s function:
∑α ( y ( ( w ⋅ x ) + b ) − 1 + ξ ) (14)
m m
1
L ( w, b, a ) = ∑
2
w + C ξi − i i i
2 i =1 i =1
Calculated:
m
∑α y
i =1
i i = 0 (15)
m
w= ∑α y x
i =1
i i i
(16)
C − α i − β i = 0 (17)
Take into the plane equation, get the decision function:
 m 
∑
f ( x ) = sgn  ai yi ( x ⋅ xi ) + b *  (11)
 i =1 
4.2. Experiment simulation

Intelligent diagnosis of rolling bearing can determine whether the normal work state
of the bearing, but also can quickly and accurately the bearing fault classification and
recognition, intelligent diagnosis model process first acquisition of bearing vibration
420 RISTI, N.º E5, 07/2016

signal, and then by wavelet decomposition and wavelet packet energy extraction, signal
energy characteristics are obtained. Then, the optimal feature vector is obtained by using
the data preprocessing with the extreme value normalization formula. Use to extract the
feature vectors to form a set of sample and the sample set is divided into two parts: training
and testing, with the training sample set to train the SVM, the SVM training and testing
samples are tested on the set, test in the diagnosis of whether arriving at an accurate rate.
Finally, using SVM to reach the requirements of the bearing can be intelligent diagnosis.
Figure 7 – support vector machine (SVM)
Bearing condition Training sample number Test sample number State mark
Normal working conditions 40 40 1
Inner fault 40 40 2
Outer ring fault 40 40 3
Rolling element fault 40 40
Table 2 – The number distribution of number
serial Spectrum 1 Spectrum 2 Spectrum 3 Spectrum 4 Spectrum 5 Spectrum 6 Spectrum 7 Spectrum 8

1 72.6214 10.5231 4.2141 2.6753 3.8264 4.1152 0.5438 1.4807
2 80.2541 7.5102 3.4126 2.5218 2.1431 3.2164 0.5211 0.4207
3 75.5287 9.2643 2.6414 3.5624 1.7815 5.3124 0.8541 1.0552
4 16.2451 5.2928 10.1475 8.9336 4.1862 8.2793 40.1627 9.7528
5 14.6953 6.7852 8.8462 6.4568 6.7852 14.5486 36.4513 5.4314
6 16.9650 4.2568 6.5482 6.2564 5.3698 10.8965 42.5623 7.1453
7 21.5634 12.6354 10.9875 24.6845 3.6984 8.3546 12.2548 5.8214
8 20.5569 15.5462 9.5246 29.3654 2.3648 4.5862 10.3647 7.9012
9 18.6754 16.4548 6.6427 30.3445 6.3544 4.7841 6.3348 10.4093
10 19.0935 38.4625 3.7505 14.8023 14.5224 7.3201 0.6550 1.3936
11 16.5736 40.4254 4.5789 12.5473 15.3784 8.4524 0.9541 1.2564
12 17.3546 39.3648 4.6951 15.3648 13.3589 7.3548 0.6581 1.4694
Table 3 – The part of training sample energy character vector
RISTI, N.º E5, 07/2016 421

First, the training set is used to train the SVM, and then use the trained SVM model
to predict the test set, in which the SVM program uses the libsvm toolbox. Figure 8
is the fractal dimension of data, which indicates that the 320 sets of feature vectors
extracted by wavelet packet are distributed in each frequency band. Figure 9 is the final
classification result. Table 4 is the statistical table for diagnostic results.
Figure 8 – Data points dimension visualization diagram
Figure 9 – The classification results diagram of test set
422 RISTI, N.º E5, 07/2016

Correct
Test sample Comprehensive
Bearing condition identification accuracy rate
number accuracy
number
Normal working
40 36 90%
conditions
Inner fault 40 40 100% 94.375%
Outer ring fault 40 37 90.25%
Rolling element fault 40 38 90.5%
Table 4 – Result of fault diagnosis

Diagnosis results show that the normal bearing 4 false positives, the inner ring fault does
not appear false positives, outer ring fault occurs 3 miscarriages of justice, the rolling
body fault occurs 2 false positives. Its diagnostic rate was 90% (36/40), 100% (40/40),
90.25% (37/40),90.5% (38/40).The final diagnosis accuracy rate is 94.375% (151/160),
which is consistent with the accuracy of bearing fault diagnosis.
5. Conclusion
This paper analyzes the failure forms of rolling bearing, fault characteristic frequency
and vibration mechanism and application of wavelet de-noising, wavelet packet energy
extraction and support vector machine (SVM) and other technologies, using a rolling
bearing fault intelligent diagnosis method based on support vector machine (SVM),
verified by simulation experiments. The main results the following: due to the bearing
signal is mixed with a lot of noise signal, the wavelet de-noising technique can effectively
remove the original signal in the noise signal, effectively reduce the after fault diagnosis
of error. Wavelet packet energy decomposition is a kind of effective of rolling bearing
vibration signal feature extraction method, it can extract the signal energy in different
frequency bands, the characteristic information of the signal in each frequency band
is composed of feature vector can be a good characterization of rolling bearings. This
article in view of the limited number of fault samples, the wavelet threshold denoising
and support vector machine combination method to diagnose the fault of rolling bearing.
Experimental simulation results show that the proposed method can identify the fault
types of rolling bearings rapidly and accurately.
Acknowledgments
The work of this paper is supported by Project of Natural Science Foundation of
Heilongjiang Province (E201215); The Ministry of education, the specialized research fund
for the doctoral program of higher education of doctoral research (20120062110006).
References
Abhijit, K., Jauaraman, V. (2005). Knowledge incorporated support vector machine to
detect faults in Tennessee Eastman Process. Computer and Chemical Engineering,
29, 2128–2133.
RISTI, N.º E5, 07/2016 423


Farshid, T., Farshid, T. (2007). A ball bearing fault diagnosis method based on wavelet
and EMD energy entropy mean. International Conference on Intelligent and
Advanced Systems, 1210–1212.
Rai, V. (2007). Bearing fault diagnosis using FFT of intrinsic mode functions in Hilbert–
Huang transform. Mechanical Systems and Signal Processing, 21 , 2607–2615.
Sugumaran, V., Ramachandran, K. (2007). Automatic rule learning using decision tree
for fuzzy classifierinfault diagnosis of roller bearing. Mechanical Systems and
Signal Processing, 21, 2237–2247.
Xiaodong, W., Yanyang, Z. (2011). Multiwavelet denoising with improved neighboring
oefficients for application on rolling bearing fault diagnosis. Mechanical Systems
and Signal rocessing, 25, 285–304.
Yunyun, Y., Wei, T. (2011). Study of Remote Bearing Fault Diagnosis Based on BP Neural
Network combination. Seventh International Conference on Natural Computation,
618–621.
424 RISTI, N.º E5, 07/2016

Research on Public Management Efficiency

Improvement Method based on Parallel Database
Oriented Optimization Management Information
System
Yingjun Han1,*, Rui Sun2
* bearboy227@163.com
North China University of Science and Technology, 063000, China

1
2
North China University of Science and Technology Affiliated Hospita, 063000, China
Pages: 425–436
Abstract: In recent years, with the rapid and bursting development of manage
information system and the data base technology, the applications have been
varying from general science and social science applications. Under this condition,
this paper proposes the novel public management efficiency improvement method
based on the parallel database oriented optimization management information
system. The history of the public administration paradigm shift shows that the
task of public management and the form is always the certain historical period
of the development level of productivity and the production organization form of
the decision and restriction. Public administration is to provide public goods to the
society. With the MIS and DB theory, we can improve the mode as these aspects
which are reflected from the MIS architecture optimization. (1) Loose coupling of
the Web Services directly from loose coupling business elements, elements of our
business in the design business process must be separated from business elements.
(2) OA is cross-platform, not be restricted by using any language and operating
system, but to the upper open API interface that provide the basic algorithm. (3)
With the data layer support, the implementation services for the entire system
provide a solid foundation. The quantitative analysis has been conducted to verify
the systematic effectiveness.
Keywords: Public management, Efficiency, Improvement method, Parallel
database, Management information system, Optimization
1. Introduction
Efficiency and its implementation mechanism have been the general management and
the core content of the traditional public administration research. With the evolution
and the development of subject, the efficiency of the increasingly rich content, the
concept of efficiency is often used on the different meaning. The resulting confusion
leads to the discussion of the efficiency of failed to further progress (Aji, 2013;
Andrews, 2013; Abreu, 2015). Neglect the uniqueness of different efficiency caused
RISTI, N.º E5, 07/2016 425

Research on Public Management Efficiency Improvement Method
by the realization of the efficiency mechanism made out prescription didn’t suit the
remedy to the case and implicit is key problem of a single research perspective (Hoi,
2013; Lannier,2014; Khitun, 2015). Public management is the basic problem of social
fairness problem is the study of social public sector how to formulate and implement
right of public policy and adopt proper management measures and regulation behavior
prompted the society as a whole to form a fair and just environment to make the
reasonable market competition rules, so as to promote the development of the society
and make of all the members of the society in various fields of social life have equal
opportunities and equal rights to get the right development(Lee, 2012; Trigo, 2012;
Osborne, 2013). The figure one illustrates the principles.
Figure 1 – The general components of the public management issues

We can be in the public management efficiency is divided into two different levels,
namely organization efficiency and development efficiency. (1) Allocative efficiency.
When talking about the relationship between the government and the market,
the mainstream view is: on the one hand, market failure and government failure
are realistic(Wang, 2014; Woodbridge, 2015). On the other hand, the government
and the market can be heals each other and this view also apply to the market
mechanism of public goods supply. (2) Organizational efficiency. Even if main body of
public administration organization efficiency is high, it provides the public goods
and public service is entirely possible from the demand of the public. In this sense,
the purpose of introducing market mechanism in public management not only with
the ability to create competitive environment, more is to give public choice and
collective judgment independently (Yfoulis, 2012; Wu, 2013).
To handle the mentioned issues, in this paper, we propose the novel public management
efficiency improvement method based on parallel database oriented optimization
management information system (MIS). Management information system is
concentrated reflection of modern management modernization and the modernization
is a comprehensive concept that is the science and the technology management system
in people’s ideology and behavior habits of modern fusion (Zhu,2015). Effective
management information system is the modern management ideas and methods,
modern organization system and advanced computer technology and the organic unity
426 RISTI, N.º E5, 07/2016

of the modern people, and that will be the latest technical and the management results
of the dynamic and the generation of the concept of dynamic.
2. The parallel database and optimization countermeasures

2.1. Parallel database concept
Query optimization in the parallel database, the processor allocation is always an
important problem, in the parallel relationship database, more join queries is the query
requirement of database application system is indispensable, and execution of each
connection operation cost depends on the processor number assigned to the connection.
As a result, more connected to the execution order scheduling and each connection
processor allocation problem is the key of query optimization in parallel library. In the
table one, we show the traditional load of the hyperbolic function.
Parameter Item number Curve function-1 Curve function-2

Q1 Xy=25 Xy=35
Selected parameter Qm Xy=95 Xy=105

core set Qh Xy=175 Xy=185
Qs Xy=355 Xy=365
Table 1 – Load of the hyperbolic function

To query optimization in distributed database, our main goal is with the minimum
total cost, in the shortest possible time to obtain required data. Its implementation
is related with the communication time already, also related to the local processing
time depending on the interconnection network condition can have different query
optimization strategies. Object-oriented database, however, is the relatively new
technology, it is lack of theoretical support, he may have in dealing with a large number
of data contains a lot of affairs much more slowly than a relational database system,
but people have developed a hybrid object database, the database will be a relational
database management system’s ability to deal with affairs and object-oriented database
systems dealing with the complex relationships and the ability of the new data. In the
figure two, demonstrate the architecture.
Optimization method is one of the research of parallel database of a part of the optimization
of the whole database query has important research value. In big ways, it can be divided
into four types, respectively is the traditional optimization method based on the query
tree and multiple weighted tree optimization method, the semantic query optimization
method, and based on the genetic optimization algorithm. Among them, the traditional
optimization method based on the query tree mainly includes the query optimization
algorithm based on linear left tree, the query optimization algorithm based on the
linear right tree, query optimization method based on linear episodic right tree, query
optimization algorithm based on the thick trees, and the query optimization algorithm
based on operation of forest which could be demonstrated as the follows, respectively.
•• A linear tree query optimization algorithm based on the left:
Schneider presents a linear tree to the left and the HASH-JOIN algorithm on
RISTI, N.º E5, 07/2016 427

Figure 2 – The architecture of the parallel database
the basis of MJ query optimization method. MJ given query Q, and Q each left
linear tree contains only one with the highest parallelism query plan, known as
Q left linear tree optimization of parallel query plan.
•• Query optimization algorithm based on the thick trees: GBT has
the very high parallelism query optimization method, but the thick tree query
execution plan space is very large. It uses the characteristics of the linear and
dense trees have very high parallelism and relative GBT smaller search space.
It can be seen that LBT tree query execution plan than GBT much smaller, and
the relationship between n increases, while this phenomenon is more obvious,
visible LBT tree based query GBT has improved a lot.
428 RISTI, N.º E5, 07/2016

•• Multiple weighted tree optimization method: This method can deal with
the choice of the most commonly used a projection of the join query, support for
multiple parallel connection algorithm, including assembly line buffer memory
allocation algorithm optimization, manipulation processor and memory
optimization allocation algorithm and the connection operation to realize
optimization algorithm of the algorithm. Query optimization algorithm that can
support variety of parallel connection algorithm provides connection operation
implementation algorithm of heuristic optimization algorithm.
2.2. Database optimization methodologies

Database optimization is to adjust the component and improve the performance,
increase the throughput of the database and the processing time is correspondingly
reduced. And database query is a management information system, the main operating
speed of the query is directly related to the performance of database system, the index
can improve the database query speed. As a result, in improving the performance of
information management system is of great significance for the optimization design
of relational database index. Running speed and the execution efficiency is an important
standard to measure the performance of the database system, in order to guarantee the
system performance and throughput, the need to optimize the database logic design and
physical design, and through the database design always, this is the first task of database
optimization design.
For this concern, we should firstly discuss the characteristics of the components.
(1) Foreign key role is to establish the relationship between the relational database
referential integrity primary key can only be from independent entity migration to the
independent entity which become a property known as foreign keys. (2) Keys are used for
complex SQL statements frequently used in the data access a table with only a primary
key. Primary key can be divided into natural primary and primary key natural primary
key by the entity attribute of the natural primary key can be a compound in the form a
composite primary key primary key columns cannot too much composite primary key
complicates Join operation also increased the size of the external key table primary key is
key in the absence of appropriate nature or nature complex or high sensitivity in human
form. (3) Index was used to optimize the system performance is obvious to all commonly
used in the where clause of the column in the query and all for sorting the columns of the
create index can avoid a full table scan or visit without changing the physical structure of
the table under the condition of direct access to specific data column.
Choosing the right column to establish the index has an important effect on the positive
and negative effects of the balanced index. Usually have the following characteristics
data column is indexed in: (1) define a primary key and foreign key columns of data;
(2) need in the specified range of fast or frequent queries in a column; (3) the need
according to the sort order fast or frequent search column; (4) in the collection process
need to fast or frequent combination to columns together.
When designing the database, the query efficiency is very important. In the design
should be noted: (1) Whether the use of index. The database index is like a directory
in the general books, when accessing the database the user can reduce the number of
data traversing the data, so as to quickly find the desired data. (2) I/O throughput. In
RISTI, N.º E5, 07/2016 429

Figure 3 – The popular adopted architecture of the database organization pattern
the hardware environment, optimization of the deployment scheme can improve the
I/O throughput and avoid the bottleneck problem. When in a certain field for query
conditions, local area need to back a large amount of data, should establish a clustered
index on this field, and when the query data quantity is small, it is necessary to establish
the clustered index in this basic field. (3) The calculation column should be created. If
there is no corresponding calculation column, in the data query on the existing data
to be calculated again, so as to waste a part of the performance. (4) To have enough
memory. In the query or access to the database in the data, it is inevitable to take up a
lot of memory space, when the memory is insufficient and the data access speed will be
significantly affected. In a large database design, therefore, only the reasonable index
field selection and the rational use of index design principles that can effectively improve
the operating efficiency of the entire database.
Figure 4 – The proposed DB optimization architecture
430 RISTI, N.º E5, 07/2016

3. Public management efficiency quantitative analysis

3.1. Quantitative analysis of public management
Public administration is a high degree of cross discipline. Its research objects, and political
science and sociology has a considerable overlap, research paradigm and management
more similarities, analysis tools are most reference to economics and statistics. Therefore,
a complete public management research methodology should be a methodology for
organic integration of related disciplines can form a scientific analysis of all the public
management thinking system. Quantitative analysis exist obvious form characteristics,
according to the characteristics and the quantitative analysis of the three forms whether
to build a hypothesis model in this paper to illustrate public management problem. Under
this basis, research methodologies can be generally separated into the following aspects.
•• To research the measurement, using statistical methods described, such as
charts, figures to illustrate this, quantitative analysis is to interpret the problem
of the auxiliary means, this kind of literature for the study of the quantitative
analysis as the auxiliary pole.
•• For the creation of the experimental conditions, control variables, collect data,
hypothesis testing for things to research abstracts a number of the variables,
structure model, the relationship between variables, structure study as causal
research.
•• In addition to the above two cases in the literature, in order to form such as
case, presentation, retort, argue to discuss the problems and get the research
conclusion, this kind of research for the pure qualitative research.
Positivism is trying to build can explain and predict the theory of nature and social
events, and to develop a set of interrelated and can verify rules, these rules explain the
causal relationship between variables, the ultimate goal is to control the natural and
social events. According to this thinking, the fact of social life or organization can be
separated from values and the function of the science is to focus on facts, not values. For
this, the table 2 shows the related principles.
Research stage Principles
Informed consent: inform the basic information about the research subjects,
Start
and obtain their consent
Privacy and confidentiality: respect the research object of personal privacy,
Data gathering and promises to its confidentiality to avoid damage, rude, and reluctantly: to
avoid harming the object of study
Anonymous: study cannot be able to identify the object of study status of
Writing reports information; Justice: description of the object of study the objective and fair
attitude
Reasonable use, ensure that the study found that to be used reasonably, and is
End of the study
used for legitimate purpose
Table 2 – Quantitative analysis of public management steps and principles

How to better carry out quantitative study of the public management, we need to clarify
the relationship between the two. One is found to the problems of research, analysis and
solution to clarify the relationship. These three relations can be simplified as the study of
RISTI, N.º E5, 07/2016 431

causal relationship between the research questions, in the face of the complicated public
affairs, public the management is the main task of researchers finds public affairs internal
causality. The causal relationship within public affairs of the essence can use mathematical
function relational expression. In the formula one, we define the relationship.
y = f ( x1 , x2 ,, xn ) (1)
The relation between highly reflects the nature of public affairs within the causality, and
the relationship can be expressed as formula two for simple expression.
y = f (x) (2)
Qualitative research with quantitative research is the important tool of the road to solve.
Qualitative research is mainly use the interview or literature that uses the narrative
techniques. In contrast, quantitative research is used the social survey data, using
statistical analysis technique. In fact, the difference between qualitative and quantitative
research method is far from imagination of big, in many cases, the difference between
them is more in the style of study and use of specific skills and technology. Both
quantitative research and qualitative research, in the process of the causal analysis of
logic and the effect is the same.
3.2. Public management efficiency

The unity of social development and social justice is the human eternal pursuit of goals
that is the eternal theme in human society. A society must also constantly development
inevitably. But in the process of social development, as a result of people between
the status and the difference of the ability and opportunity, social differentiation and
inequality are continually. This leads people to the pursuit of social justice and voice,
and also causes the social conflict again and again. But in the class society before the
market economy, the social order is rely on force to maintain the rule, social member’s
possession of social wealth is subject to their status and rank, social justice is only a
fantasy and is in a state of the slow social development. To enhance the efficiency of the
organization depends on internal resources configuration status and level of internal
management. Both in nature can be summed up in the organization management
system and the management pattern of the results. In comparison, the third sector as
the main body of public management organization system is not as large and complex as
the government. Many government departments, as different levels of government are
government internal management must face, with this condition, we should consider the
related countermeasures as the follows. (1) Decentralization. Decentralization has two
meanings: one refers to central government to local government decentralization, local
governments, first of all, to be responsible for their jurisdiction of providing public goods,
the second is to government departments at all levels of government decentralization,
let each department is responsible for the different nature of basic public goods. (2)
Performance of the contract or performance of the agreement. Despite the adaptive
method of the private sector in the public sector by some skepticism, but do need in public
432 RISTI, N.º E5, 07/2016

sector is responsible for, and higher efficiency. (3) Building the competitive mechanism
of the public administration. To break the monopoly and encourage competition, the
concern of the public choice has been to most people. In short, that is, within the public
management subject system to introduce market competition mechanisms. (4) Build
public management incentive mechanism. Incentive mechanism to solve the problem
there are two main: for the public sector and non-profit organizations, the core of
incentive mechanism is the priority principle, according to the budget by public budget
or the public spending encourages them to provide better, more public goods and public
services. (5) The competition within the public management subject system there may
be such a dilemma: participate in the competition of subject too little is not limited and
too much involved in the main body can cause repetitive construction and waste of the
resources, as resulting in inefficient. It seems that the core role of the government should
get again and that will also be faced with the sufficient issues.
The new public management from public choice theory basis, think the government
should take the customers and the market as the guidance, which changes the traditional
public administration mode of the relationship between the government and society,
the functions of the government and its relationship with social the location again. The
new public administration thinks that the government’s social responsibility mainly is
to provide service to customers according to requirements of the customers. The idea
embodied in the new public management that put some scientific management methods,
such as management by objectives, performance evaluation, cost accounting is the
public administration field, such as to improve the government’s work efficiency has a
promoting effect. This requires government to establish new management mechanism to
strengthen government management efficiency, improve the administrative efficiency.
At the same time, the new public management believes that the government should pay
attention to the efficiency and quality of management activities, that is to pay attention
to the performance of work should pay attention to analysis of input and output ratio,
Figure 5 – The public management efficiency enhancement systematic architecture
RISTI, N.º E5, 07/2016 433

the government also should pay attention to cost accounting and also must pay attention
to save administrative cost. For this goal, we should introduce the management system
as the figure 5, and we will discuss the building of the MIS system in the later section.
4. Management information system implementation

Management information system of the evaluation system of building is a process of
system, evaluation system, to build the information system must be explicitly the
basic elements of basic information system evaluation system. When evaluating an
information system, the most important is to establish the evaluation index system.
The evaluation index system includes both the information system development,
that also is the information system of the main body, also includes the direct user of
information system is the object of information system, includes more the external
social environment. Based on this principle, in figure 6, we show the architecture of
the proposed methodology. Isolation system design, the layered structure changed,
simplifies the complex degree, so the development of system is more convenient. For
better demonstration, the system is described as the follows.
Figure 6 – The proposed management information system
•• The business layer, the realization of the process: Loose coupling of

the Web Services directly from loose coupling business elements, elements
of our business in the design business process must be separated from business
elements, appropriate to the needs of the business division loosely-coupled
business elements.
•• Web service layer, business elements of basic encapsulation:
The development of the Web service layer is in the following services under
the premise of the protocol stack. Web service is actually the business layer
of the business elements to meet the requirements of the Web service protocol
stack encapsulation.
•• Rule layer, and the realization of the algorithm: OA is cross-platform,
not be restricted by using any language and operating system, but to the upper
open API interface, provide the basic algorithm.
•• The data layer, the foundation of the “service”: The data layer is
responsible for a variety of community management data stored in database,
the database management system to complete the data management. With the
434 RISTI, N.º E5, 07/2016

data layer support, the implementation services for the entire system provide a
solid foundation.
MIS project plan can be a full and also can be periodic or subsystem. With overall
planning should be as much as possible to leaving allowance and elastic; While phased
plan or subsystems can be according to the recent fine, forward general method to carry
out, so that we can make the development team fruitful work, and with the development
of the project, make customers more and more requirements and expectations on the
late specific and clear, phased plan and subsystems can be subsequently refinement,
and the entire plan can maintain stable in general. Foe the MIS assisted public
management efficiency enhancement and evaluation task, we suggest to finalize based
on the following aspects. (1) The mathematical model of basic fuzzy evaluation. Fuzzy
comprehensive evaluation method for management information system evaluation
index system is difficult to use mathematical tools of uncertain factors in comprehensive
evaluation, the evaluation index system and fuzzy mathematics theory is applied to
implement the management information system evaluation in the process of the fuzzy
representation, judgment, reasoning process, and considering the influence of various
factors to make the evaluation results more close to actual. (2) Management information
system based on neural network evaluation model is set up. The application of the
model and algorithm of neural network in information system, evaluation system of
economic evaluation based on neural network model is established, in order to more
objectively evaluate the system performance, function and the economic benefit. (3)
AHP evaluation mathematical model is set up, application of analytic hierarchy process
(AHP) to evaluate semi-structured factors in the management information system and
to establish performance management system and quality of AHP evaluation system. (4)
Final verification on the application of public management efficiency task.
5. Conclusions
Public management is the historical process of human public management practice in
people’s ideological theory. It is always influenced by the development level of social
production development. The history of public administration paradigm shift shows that
the task of public management and the form is always the certain historical period of
the development level of the productivity and production organization form of decision
and restriction. Public administration is to provide public goods to the society. To the
society in order to effectively provide public goods, public management must solve for
whom what public goods provision of public goods and provide the two basic questions
of public management. Under this theoretical background, this paper proposes the new
public management efficiency improvement method based on parallel database oriented
optimization management information system. MIS project plan can be a full and also
can be periodic or subsystem. With overall planning should be as much as possible to
leaving allowance and elastic, while phased plan or subsystems can be according to
the recent fine. Moreover, we integrate the quantitative analysis model to conduct the
systematic research on the public management efficiency which will provide the theory
with in-depth and convincing background. In the future research, we will conduct more
survey to verify the effectiveness of the proposed method.
RISTI, N.º E5, 07/2016 435

References
Informação, (16), 108–128.
Aji, A., Wang, F., Vo, H., Lee, R., Liu, Q., Zhang, X., & Saltz, J. (2013). Hadoop GIS: a
high performance spatial data warehousing system over mapreduce. Proceedings of
the VLDB Endowment, 6(11), 1009–1020.
Andrews, R., & Van de Walle, S. (2013). New public management and citizens’
perceptions of local service efficiency, responsiveness, equity and effectiveness.
Public Management Review, 15(5), 762–783.
Hoi, S. C., Jin, R., Zhao, P., & Yang, T. (2013). Online multiple kernel classification.
Machine Learning, 90(2), 289–316.
Khitun, A. (2015). Parallel database search and prime factorization with magnonic
holographic memory devices. Journal of Applied Physics, 118(24), 243–905.
Lannier, A. L., & Porcher, S. (2014). Efficiency in the public and private French water
utilities: prospects for benchmarking. Applied Economics, 46(5), 556–572.
Lee, S. K., & Yu, J. H. (2012). Success model of project management information system
in construction. Automation in construction, 25, 82–93.
Osborne, S. P., Radnor, Z., & Nasi, G. (2013). A new theory for public service management?
Toward a (public) service-dominant approach. The American Review of Public
Administration, 43(2), 135–158.
Trigo, J. D., Martínez, I., Alesanco, A., Kollmann, A., Escayola, J., Hayn, D. & García, J.
(2012). An integrated healthcare information system for end-to-end standardized
exchange and homogeneous management of digital ECG formats. IEEE Transactions
on Information Technology in Biomedicine, 16(4), 518–529.
Wang, H., & Wang, J. (2014). An effective image representation method using kernel
classification. In 2014 IEEE 26th International Conference on Tools with Artificial
Intelligence, 853–858.
Woodbridge, D. M. K., Wilson, A. T., Rintoul, M. D., & Goldstein, R. H. (2015). Time series
discord detection in medical data using a parallel relational database. In Bioinformatics
and Biomedicine (BIBM), 2015 IEEE International Conference, 1420–1426.
Wu, Y., Cegielski, C. G., Hazen, B. T., & Hall, D. J. (2013). Cloud computing in support
of supply chain information system infrastructure: understanding when to go to the
cloud. Journal of Supply Chain Management, 49(3), 25–41.
Yfoulis, C. A., & Gounaris, A. (2012). Online load balancing in parallel database queries
with model predictive control. In Data Engineering Workshops (ICDEW) 2012
IEEE 28th International Conference, 269–274.
Zhu, C., & Gao, D. (2015). Improved multi-kernel classification machine with Nyström
approximation technique. Pattern Recognition, 48(4), 1490–1509.
436 RISTI, N.º E5, 07/2016

Adjustment and Application of Ground Motion Input

in Temporal Analysis of Structure Aseismic
Changhao Zhang*, Jianhu Feng, Xuntao Wang
* zch0825@sina.com
College of Science, Chang’an University, Xi’an 710064, Shaanxi,China

Pages: 437–446
Abstract: The input of ground motion in earthquake transient dynamic analysis has
been a hot- point subject in the earthquake engineering research. Its primary problem
is the input of seismic motion. In this paper, through using the good time-fre¬quency
localization nature of the wavelet function to analyze the real ground motion, a feasible
method based on wavelet analysis is proposed. The actual seismic of ELCentro and
TH2TG055 are input and the modulated and transformed seismic of ELCentro and
TH2TG055 are adjusted by the method to the structural model. In the process of
seismic response analysis, this paper analyzed the results of storey shear force,inter
layer displacement, and storey drift angle and so on, which was indicated that the
transformed seismic waves could obtain the relative better result, and meet the needs of
earthquake resistant engineering. It has certain value in engineering application.
Keywords: Seismic acceleration; high-rise RC corewall structure; ground motion
input; security analysis
1. Introduction
The earthquake means crustal vibration caused by the rapid release of energy in the
process; earthquake related engineering structure seismic research has been the focus of
civil engineering research. Among them, choosing appropriate ground motion input is the
first consideration in the study of dynamic response and seismic design of structures, and is
also one of the main content of response spectrum method and seismic time history analysis
method(Jie, 2013; Changhao, 2012). At present, based on the response spectrum theory
and the random theory and other methods, it is more concerned with the non-stationary
phase, the phase change, the excellent frequency and the maximum amplitude of the
ground motion intensity(Ming, 2007; Lewei, 2009; Abreu, 2015). Due to the randomness
and uncertainty of natural ground motion, there is little research on the time frequency
characteristics of ground motion, which is the reason why the artificial seismic wave is not
consistent with the natural seismic wave. In this paper, wavelet analysis is used to consider
the non-stationary, phase, amplitude and time-frequency characteristics of ground motion
to adjust the actual ground motion records for seismic time history analysis(Mukherjee,
2002; Qianli, 2008). Results show that wavelet analysis method will be adjusted after the
earthquake wave used in high-rise structure dynamic history analysis, comparison of high-
rise structure adjustment before and after shock wave seismic response results, to meet the
RISTI, N.º E5, 07/2016 437

Adjustment and Application of Ground Motion Input in Temporal Analysis of Structure Aseismic
requirements of the “code for seismic design of buildings”, the structure of seismic dynamic
response and seismic design provides new methods and ideas.
Wavelet analysis method can combine the time domain and frequency domain analysis,
to describe seismic wave signal time frequency localization properties. MATLAB has a
powerful algorithm development, data visualization and analysis and other functions. In
the selection of a double orthogonality, compact support, continuous (discrete wavelet
transform characteristics of wavelet function-Daubechies (DBN) application to seismic
waves of orthogonal decomposition, divides the frequency domain for different adjacent
band, in different frequency bands to adjust and standard design response spectra were
fitted (Suarez, 2005; Zuozhou, 2015).
This paper is based on the wavelet analysis technique to adjust the actual ground motion:
1. Select the true vibration record: Reasonable ground motion input is a necessary
condition to ensure the reasonable results of structural seismic time history
analysis. Selected ground motion records the basic principle is really is the
ground motion response spectrum to and site of the proposed project standard
response spectrum is consistent; the second is the actual ground motion of the
three elements close to the norms of the vibration parameters.
2. According to the provisions of the peak acceleration of the “anti-regulation”, the
ratio method is adopted to adjust the amplitude of the acceleration of the actual
ground motion:
′
Amax
α ′ (t ) = a (t ) (1)
A max
In this formula, A max and a(t) a(t) are the absolute values and the acceleration
time history of the maximum time history of natural seismic waves; Amax ′ and
α ′ ( t ) are the specification of the different seismic intensity of the maximum
design earthquake acceleration and amplitude modulation after the acceleration
of ground motion.
3. Using Nigam method to calculate ground motion response spectrum, Nigam
method is that in all of the computation process without introducing any
approximate calculation method, will not produce any truncation and roundoff
error and has a high precision, so in the current commonly used this exact solution
of Nigam method for general processing seismic engineering earthquake record.
4. Comparison of ground motion response spectrum and design response spectrum,
In the [0.1, Tg] platform and the structure of the basic cycle T1 near [ T1 − ∆T1 , T1 + ∆T2 ]
platform segment, The mean value of the acceleration response spectrum
of ground motion control and the design response spectrum of the proposed
engineering site are within 10%, and the mean value of the response spectrum
as ∆T1 ≤ ∆T2 =0.5s is suitable.
5. To solve the k1 , k2
k 1 = Tg / Tg′ , k 2 = Tg / Tg ′ (2)
Tg′ Mean (area) of ground motion response spectrum in [0.1, Tg ]
438 RISTI, N.º E5, 07/2016

Tg Design response spectra in the vicinity of the basic structure T1 of the

structure[ T1 − ∆T1 , T1 + ∆T2 ];
T ′ Ground motion response spectrum in the vicinity of the basic structure of
g
the structure[ T1 − ∆T1 , T1 + ∆T2 ]segment of the mean (area);
6. Wavelet function is used to decompose the amplitude modulated seismic wave
in frequency domain, Then each frequency domain is divided into adjacent
frequency bands, the corresponding frequency bands and k1, k2 phase is
multiplied by the mean of the two platforms to achieve the purpose of fitting
the standard response spectrum. And then re combine the component to form a
adjusted seismic wave, analysis of the peak acceleration to meet the specification
requirements; if not meet the requirements, the need to re adjust to meet the
requirements of seismic waves.
2. Ground motion wavelet analysis

2.1. Actual vibration selection
Follow the steps mentioned above, this paper selects El Centro wave and TH2TG055
wave. According to the proposed project site seismic fortification intensity is 6 degrees,
site classification is II, earthquake are grouped in the third group, the earthquake
acceleration time history the biggest value according to table 1 selected. The ratio
method is used to adjust the acceleration amplitude of the actual ground motion, and
the adjustment of the wavelet transform under the rare earthquake is carried out.
Earthquake effect 6 degrees 7 degrees 8 degrees 9 degrees
Multi event earthquake 18 35(55) 70(110) 140

Rare earthquake 125 220(310) 400(510) 620
Table 1 – PGA in Time history analysis
2.2. Wavelet transform and comparison of ground motion

Figure 1 (a) (b) is the amplitude modulated Elcentro wave under earthquake and
compared with the design response spectrum. According to figure 1 (b), we can know
that the two are in the[0.1,Tg]([0.1,0.45]) platform segment and [ T1 − ∆T1 , T1 + ∆T2 ]
([0.5,1.5]) segment gap is larger, so the use of wavelet transform to adjust the Elcentro
wave fitting design response spectrum. By using Daubechies (db3) to decompose the
Elcentro wave, the wave is decomposed into 6 frequency bands. Reconstruction of each
layer by using the reconstruction command wrcoef, as shown in Figure 1 (C), Figure 1
(d) is the use of the various layers of the Fu Liye transform, the spectral density curve
of each layer. Then use (1.2) to obtain the response spectrum [0.1,Tg] and [ T1 − ∆T1 ,
T1 + ∆T2 ] platform section of the area of the adjustment coefficient, the adjustment
coefficient is multiplied by the frequency band of small amplitude value. Figure 2 (a) is
re combination of each component is formed after adjustment for seismic wave and the
adjustment of seismic wave can be seen using wavelet analysis method to adjust seismic
waves can meet the requirements of specification. According to the method of Table 1
RISTI, N.º E5, 07/2016 439

and above, the amplitude modulation and wavelet transform of ELCentro wave under
rare earthquake are shown in Figure 2 (b), and the fitting error is less than 10%.
Figure 1 – Modulated and transformed process of El-Centro wave in frequently occurred

earthquake(a-d)
In order to verify the rationality of the method and to compare the input high level
structure, Figure 3 is the TH2TG055 wave after am and wavelet transform, the results
obtained to meet the requirements, can be used as a structural time history analysis of
the input ground motion records.
Figure 2 – Modulated and transformed comparison of El-Centro wave in frequently and rarely
occurred earthquake
440 RISTI, N.º E5, 07/2016

Figure 3 – Modulated and transformed comparison of TH2TG055 wave in frequently and rarely
occurred earthquake
3. Ground motion input high level structure analysis

3.1. Structure system introduction
Ji’nan city commercial complex project of a plot by the senior The Residence Hotel,
business building and multi-storey commercial podium, the total construction area is
about 28 million square meters. The overall height of the business integrated tower is
150.45m, the ground floor 34, with 3 floors of the basement; comprehensive building
standards for 4.1m, the bottom four layers of high are 4.5m;The plane is rectangular,
and the length and width of the typical floor are 54.8m and 36.8m respectively. Tower
frame - core tube as the anti-lateral force system, the core wall thickness of the outer wall
of 700mm~400mm, the internal wall thickness of 300mm~250mm.The structure of the
third layer and the bottom of the frame column section size is 1250mmx1250mm, the fourth
layer to the fifteen layer of the frame column section size becomes 1200mmx1200mm,
1150mmx1150mm. The height of each frame beam is 850mm~700mm. Beam and slab
concrete grade is C30, column, wall concrete grade is ~C60 C40.
3.2. Establishment and calculation of the model

Projects seismic fortification intensity is 6 degrees, the acceleration 0.05g, earthquake
are grouped into the three groups, site class II, cycle characteristics =0.45s and
structural damping ratio of 0.05, the basic wind load standard value (according to the
wind pressure once in 50 years). Overall structure were calculated using SATWE, rare
earthquake is calculated by using the ABAQUS software and take the above adjustment
before and after shock wave, considering the effect of elasto-plastic dynamic time history
analysis. The structure calculation model is shown in Figure 4.
This paper, taking 15 modes is analyzed, considering the effect of torsion coupling
and accidental eccentricity and clear structural mode of vibration, good self-vibration
characteristics. The 1 translational cycle T1=4.71s (Y direction translation), the cycle
than Tt/T1=0.76; structure in X and y to the maximum displacement and the ratio of
average displacement maximum value of 1.13; maximum story drift angle is 1/1105 in
the 23rd floor, specification of frame core tube is less than the interlayer displacement
angle maximum 1 / 800; structure the stiffness ratio of 1.13, on the twelfth floor of the
structure, minimum floor shear bearing capacity of 0.86, vertical structure rules, there
RISTI, N.º E5, 07/2016 441

is no weak layer. The overall stability of the structure is more than 1.4 but less than 2.7,
which indicates that the structural system is stable, but it needs to consider the adverse
effects of the two order of gravity. The other indexes all meet the requirements of the
standard, and the plane layout of the structure is reasonable and effective.
Figure 4 – Structure calculation model and Finite element model
3.3. The ground motion input to the high-level structure analysis

The dynamic time history analysis to select the second section wavelet transform
adjustment before and after the actual vibration - El Centro wave (E wave) and
TH2TG055 wave (referred to as H wave), the effective duration of not less than 5 times
of fundamental period of the structure; seismic wave by two-way input and primary and
secondary direction and speed of the maximum value of the ratio is 1: the conforms to
the proposed site often encounter and seldom occurred earthquake requirements.
Figure 5 is in the case of earthquake, before and after the transformation, the E wave,
H wave input high level structure elastic time history analysis of the main direction of
calculation results. From Figure 5 (a) that each of the seismic base shear values were
greater than the mode decomposition response spectrum method to calculate the results
of 65%, and the calculation results are not more than 135%; structure base shear average
value is greater than that of vibration mode decomposition reaction spectrum 80% of the
bottom shear method, selected after adjustment of seismic wave input structure elastic
history analysis meet the specification requirements. The interlayer displacement map
seen through the transformation of the elastic seismic history analysis results and mean
response spectrum calculation results are slightly different, two value 3.90% difference,
and transform of seismic wave results mean and response spectrum calculation results
the difference between the value of 5.51%; before the adjustment, after seismic wave top
layer displacement dispersion of a wave and response spectrum analysis get the top floor
displacement respectively, a difference of 15.4% and 12.6%. By comparing Figure 5 (c)
442 RISTI, N.º E5, 07/2016

and elastic time history analysis and response spectrum method calculated interlayer
displacement angle did not exceed the allowable value of drift angle of 1 / 800, can meet
the requirements of specification and transform of seismic wave interlayer displacement
angle value, the overall to response spectrum curve near.
From Figure 5 shows that using wavelet transform of seismic wave analysis as elastic
structure history analysis input, numerical calculation results and the response spectrum
method is closer, to calculate the index can meet the requirements of specification, the
method for seismic design of engineering provides a new way.
Figure 5 – Elastic time history analysis results of structure in frequently occurred earthquake
With the continuous improvement of the structural dynamic model and the increase of
the efficiency of numerical calculation, the nonlinear dynamic analysis of structures is
becoming more and more mature, and it is more and more widely used in the seismic
design and performance evaluation of complex structures. Dynamic elastic-plastic time
history analysis is an effective method to predict the seismic response of structures and
to evaluate the seismic performance of structures. This paper uses ABAQUS for high-
rise structure rare met under the action of seismic elastoplastic time history analysis,
the figure 2 and figure 3 am and transform the before and after four wave as input wave,
before and after the study of wavelet transform to analyze the actual ground motion
on the dynamic response of the difference. The model wall concrete constitutive using
ABAQUS comes with the damage model of concrete, the concrete beam constitutive using
fiber beam model, steel using bilinear kinematic hardening model; panel unit using S4R
multi-layer shell element [-], the beam element with B31 unit. The time history analysis is
carried out using the display algorithm, and the calculation results are shown in Figure 6.
Rare earthquake is about frequent earthquake and earthquake design level is 7 times
and 2.33 times, but from Figure 6 (a) results can be seen rare earthquake and frequent
earthquake at the bottom of the structure seismic shear than average is 4 to 5 times,
indicating that the component by nonlinear stiffness degradation to reduce earthquake
in. Under rare earthquake, the structure can still be kept upright, and the maximum
displacement angle of the main direction is 1/294, which is less than the limit value
of the norm 1/100. Table 2 shows, as the structural elastoplastic time history analysis
input adjustment before, after El Centro wave and TH2TG055 wave of base shear and
RISTI, N.º E5, 07/2016 443

name ELCentro wave TH2TG055wave

Relative Relative
category front back front back
difference% difference%
Base shear 51454 45150 14.2 28983 31493 8.7
Top displacement 342.3 323.7 5.7 252.4 220.4 14.5
Maximum drift
angle between 1/294 1/322 9.5 1/408 1/426 4.4
layers
Table 2 – The comparison of pre and post wave analysis results in ground motion
Figure 6 – Nonlinear time history analysis results of structure in rarely occurred earthquake
444 RISTI, N.º E5, 07/2016

top displacement, the maximum displacement angle value difference of not more than
20%, in the statistical significance conform to the rules and regulations, identified the
selection of earthquake wave effectively. Figure 3.4 (d) is extracted from the structure
of the tenth layer of the structure of the internal force time history curve, respectively,
when the 2.7s and 5S, the maximum value of internal forces; also, from Figure 6 (E)
shows that the maximum value of the displacement angle in 2.3S and 5.8s. From Figure
6 (a) - (E) can be seen, the use of wavelet transforms before and after the 2 kinds of
discrete wave of smaller, the curve is smooth and consistent.
4. Conclusions
This article in detail introduced based on the analysis of wavelet transform method to
adjust the actual ground motion process, select different periods of different frequency
bands reconstruction, re combination of components that match the target response
spectrum form adjustment after the earthquake wave, achieve adjusted the vibration
frequency nonstationarity. In engineering practice, selection based on the wavelet
analysis method in the adjustment of seismic wave as a structural time history analysis
input, through comparison of interlinear shear, the interlayer displacement, interstory
displacement angle and so on several aspects of the calculation results can get more
accurate numerical value, meet the seismic design needs, in the engineering application
has a certain reference value to the application and further structural seismic design
improvement and perfection to provide a new way.
References
Informação, (16), 108–128.
Changhao, Z. (2012). Simulate the Seismic Wave and Analysis Dynamic Response of
Engineering Structures. Xi’an Chang’an University, 12–15.
Jie, C. (2013). The Research of Earthquake Ground Motion Input in High-level Structure
Adjustment Based on Wavelet. Xi’an University of Architecture and Technology,
20,13–15.
Lewei, M., Xiao-jun, Z. (2009). Dynamic Characteristics Analysis Method of The
earthquake waves on The basis of Matlab language. Journal of Shaanxi University
of Science & Technology, 27, 130–134.
Ming, Z. (2007).Inelastic and failure analysis of laminate structure by ABAQUS
incorporated with a general constitutive relationship. Journal of Reinforced Plastics
and Composites, 26, 1135–1181.
Mukherjee, S., Gupta, V. (2002). Wavelet-based generation of spectrum-compatible
time-listories. Soil Dynamics and Earthquake Engineering, 22, 799–804.
RISTI, N.º E5, 07/2016 445

Qianli, M. (2008). Study on lateral load patterns of pushover analysis using incremental
ynamical analysis for RC frame structures. Journal of Building Structures, 2,
132–140.
Suarez, L., Montejo, L. (2005). Generation of artificial earthquakes via the wavelet
transform. Journal of Solids and Structures, 42, 5905–5919.
Zuozhou, Z. (2015). Comparison of ground motion selection between Chinese and
American methods and elasto-plastic time history analysis of frame-core wall
structures. Journal of Building Structures, 36, 10–18.
446 RISTI, N.º E5, 07/2016

Design and Development of a Dynamic Wearable

Computer Device Platform
Zhouping Yin*
*Yinzhouping@aqnu.edu.cn
Anqing Normal University, Anqing 246001, China

Pages: 447–458
Abstract: In this paper, the design of wearable computer system is carried out by
using the software and hardware coordination method, using the ISE9.1, EDK9.1
and PlanAhead9.2.7 tools to complete the design of the entire system. The wearable
computer system design is divided into several parts, and each part of the system is
divided into several modules. The static module will not change during the operation
of the system, and the dynamic module function will change with the change of
the functional requirements, and will be re configured in the system operation
process. Through the test and analysis of the performance of the configuration, the
correctness and efficiency of the proposed design method is verified.
Keywords: Wearable computer equipment, SOPC, FPGA, hardware design,
reconfigurable computing
1. Introduction
In this paper, the design and implementation of a wearable computer dynamic
reconfiguration module based on SOPC is studied. Currently on the market can be
wearable computers are based on COTS technology, the traditional design idea is the
different combination of the functional circuit, according to the application demand to
match circuit, to achieve different functions, the scale of such a system is the sum of all the
function modules(Asim, 2012; Fang, 2014). With the increase of system requirements,
the scale of the system is expanding, the circuit becomes more complex while power
consumption is more and more big, and it faces the design bottleneck of volume and
power consumption (Becker, 2002; Lysaght, 2006). Therefore, the design goal of the
new wearable computer is small size, low power consumption, at the same time, it should
take care of flexibility; can improve the utilization of resources, with limited hardware
resources to achieve more flexible. With the development of VLSI (very large scale
integrated circuit, and has the mass of logic resources of programmable logic device,
system on chip (system on a chip (SOC) the rapid development of and programmable
logic device FPGA can already be achieved previously only ASIC (application specific
integrated circuit) chips can complete functions in today’s mainstream, and based on the
SOC design is more flexible. In this development trend, many FPGA vendors provide a
solution to SOC, a Programmable (System Chip SOPC) can be programmed on the chip
RISTI, N.º E5, 07/2016 447

Design and Development of a Dynamic Wearable Computer Device Platform
(on). SOPC can be embedded in a system to a single chip FPGA, in the above to achieve
the logical function of the entire system. The design idea of SOPC accords with the design
idea of the new wearable computer. Using SOPC technology can make the wearable
computer smaller and has the advantage of COTS technology (Sá, 2012; Zhu, 2014).
Dynamic reconfiguration technology is used to meet the requirements of high flexibility
and low power consumption of wearable computer. By re configuring the logical resources
that can be specified by the wearable computer reconstruction module, the corresponding
logical functions are completed. At the same time, it does not affect the normal operation of
the wearable computer fixed module, which not only improves the utilization of resources,
but also reduces the power consumption. Dynamic reconfiguration technology as an
emerging technology, combined with the hardware efficiency and software flexibility and
is one of the main techniques widely used in FPGA development at present, is also in this
paper, the design of model can be wearable computer with the main technology.
2. SOPC design method and partial reconfigurable technology

2.1. Development of SOPC based on FPGA Xilinx
SOC is generally used to achieve the ASIC, and the current development is very
rapid SOPC began to become a major solution for SOC. SOPC obvious advantages;
it can be completed by a single FPGA chip the main logic function of the whole
circuit system. In this paper, according to the special application background of
wearable computer, the embedded CPU core and peripheral IP core are reduced,
and the function module of the system is re configured. This design method not
only improves the utilization of FPGA internal logic resources, avoid the waste of
resources, but also make the design more flexible. SOPC has a very broad application
prospects, it occupies a more and more important position in the electronic design.
In this paper, the design method of SOPC is used in wearable computer. Compared
with the traditional COTS technology, the design of the system can be realized in this
paper. It can significantly reduce the volume of the system. This is the reason why
we choose this technology.
Figure 1 – The SOPC system
448 RISTI, N.º E5, 07/2016

FPGA is a field programmable gate array, is further developed on the basis of PAL, CAL,
CPLD and other programmable logic devices, and is the most widely used reconfigurable
logic device. FPGA chip needs to be repeated programming, it cannot be like ASIC,
through the gate circuit to achieve the basic functions of combinational logic, and can
only be achieved through an easy to configure the structure. Lookup table (Look-Up-
Table, LUT) has a structure that is easy to configure repeatedly. FPGA chips are used in
the mainstream of the SRAM process to find the table structure. SRAM FPGA chip has
the volatile characteristics, power will be lost the original configuration information, so
every time after power need from the external nonvolatile memory in the file to import
the configuration used in this paper is a Xilinx virtex-4 FPGA. The FPGA SRAM based
lookup table structure, with three LX, SX, FX platform, integration of up to 20 million
logic elements, with up to 500MHz performance and unparalleled system characteristics.
The development board is based on FX Virtex-4 platform; its main chip is xc4vfx60, the
basic internal structure of the FPGA chip as show in figure 2.
1. IOB: IOB is an input and output unit, which is a part of the interface between
the FPGA chip and the external circuit. In order to facilitate the management
and to adapt to a variety of electrical standards and iob FPGA is divided into
several groups, each group of standard interface is determined by the voltage of
the interface, a group can only have a voltage interface. The interface voltage but
different groups may be different. Only the same electrical standard port can be
connected together.
2. CLB: CLB is a configurable logic block, which is the most basic logic unit in
FPGA. CLB’s Virtex-4 consists of 4 identical Slice and additional logic. Each CLB
module can not only be used to implement combinational logic and sequential
logic, but also can be configured for distributed RAM and distributed ROM.
3. DCM: DCM is a digital clock management module. Virtex-4 FPGA Xilinx
provides digital clock management and phase loop lock. Phase loop lock can
provide accurate clock synthesis, and can reduce jitter, and achieve the filtering
function.
4. BRAM: Xilinx most of the FPGA are provided with embedded block RAM
memory, which greatly expanded the scope and flexibility of the application
of FPGA. Block RAM can be configured as a single port RAM, dual port RAM,
content address memory (CAM), as well as FIFO and other commonly used
storage structure.
5. Underlying embedded function unit: Embedded function module mainly
refers to the DLL (delay locked loop, PLL (phase locked loop), DSP and the
MicroBlaze soft core processor. Now more and more embedded functional
units, so that it has the ability to combine the software and hardware design, so
that designers can be on a single chip FPGA system level design.
6. Special embedded hard core: The major FPGA suppliers in the chip
dedicated to improve the performance of the FPGA core. Xilinx company not
only implanted in the Virtex-4 series IBM PowerPC FPGA processor, but also
developed a system level design tools EDK. Based on Miroblaze, IP and other
PowerPC core, can be carried out with the on Programmble system (System
SOPC Chip), the relevant research.
RISTI, N.º E5, 07/2016 449

Figure 2 – The internal structure of Xilinx the virtex-4 chip
2.2. Reconfigurable technology

Since the introduction of the first Xilinx support reconfigurable FPGA, reconfigurable
technology has gradually been valued by people. Dynamic reconfiguration design
method has great flexibility, for example: the use of dynamic reconfiguration, according
to the need to load the corresponding peripherals, and with different functions of the
circuit, in order to achieve different applications. When a function needs to be replaced,
without dynamic unloading of peripherals, not only to save resources, but also reduce
the power consumption. And there was no need for reprogramming of the device, as
long as to change the corresponding logic unit. In the re programming of the FPGA
internal reconfigurable module, the fixed module circuit has been working to continue
running, and will not be affected by the dynamic reconfiguration operation.
According to the time of the reconstruction operation and the behavior difference of the
reconstruction process, the partial reconfigurable technology can be divided into static
reconfiguration and dynamic reconfiguration. Static reconfiguration is one of the most
simple and widely used reconfigurable technology, which is characterized by the re generation
of a global configuration file, and then downloaded to the FPGA. After configuring FPGA, the
specified operation task is performed, and during this period, the logic function circuit of
the FPGA is kept unchanged. The feature of dynamic reconfiguration is that it can
Figure 3 – Reconfigurable Computing
450 RISTI, N.º E5, 07/2016

reconstruct the reconfigurable logic module on the FPGA at the same time of the normal
operation of the FPGA logic circuit. The more important point is that the circuit of
the fixed module cannot be affected by the normal operation when the reconfigurable
module is reconstructed. According to the different logic function of the reconfigurable
module, a number of different profiles will be generated, and then the configuration file
corresponding to the logical function can be loaded on the FPGA. In the implementation
process of a variety of logical functions need to be repeatedly reconstructing FPGA.
3. The hardware platform design of wearable computer

Wearable computer hardware platform is based on the FPGA Xilinx development board
design, it is a complete embedded system. The core of the hardware platform for FPGA
processor PowerPC 405, connected by a IBMCoreConnect bus and all kinds of peripheral
IP core. In this paper, the design of the IP core is provided by Xilinx GPIO nucleus,
UARTLITE nucleus, HWICAP core, DDR2 nuclear and systemace nuclear. In the
EDK embedded development kit, we can according to the needs of wearable computer
application, on the basis of hard core processor adding peripheral IP core, to build a
SOPC hardware platform, this paper designed the LCD IP core and VGAIP nuclear, for
display system related character information and wearable computer system to collect
the pictures or video information. First in the ML410 development platform to build a
PowerPC405 based SOPC hardware system, in the FPGA implementation of the internal
system processor management, bus architecture, peripheral interface, application
software storage and other functional modules. Each function module exists in the
form of IP and is connected with the FPGA core. Based on the special application of
wearable computer, this paper added the IP core has GPIO, LCD, ACE System, etc., and
other functional modules to be extended later. In the dynamic reconfiguration process
based on EAPR, the IP kernel function modules are divided into fixed modules and
reconfigurable modules. In this paper, the design of wearable computer SOPC hardware
platform framework, as shown in the following figure:
Figure 4 – SOPC hardware platform architecture
RISTI, N.º E5, 07/2016 451

Where the OPB _SOCKET core is used to connect the DCR bus and the reconfigurable
module, the internal structure of the OPB _SOCKET core is shown in the figure 5.
In FPGA Xilinx, high-end embedded microprocessor, and the microprocessor core
IBM PowerPC405 soft core MicroBlaze. IBM PowerPC 405 hard core in all aspects
of performance are superior to those of MicroBlaze soft core, can meet the needs of
wearable computer good processing performance, so in this paper, the design of choice
the hardcore of IBM powerpc405 processor as the system rather than core. Powerpc405
processor hardcore is IBM specifically for Xilinx development of processor product.
It uses is reduced instruction set computing architecture (RISC) in FPGA chip factory
has been implanted into the chip. In the Xilinx FPGA chip in the series, the only virtex
II series and virtex-4 series of certain types of chips only support, the development
platform of the main chip is virtex-4 series of xc4vfx60 FPGA, in which built two IBM P
owerPC405 hardcore, performance is enhanced.
Figure 5 – OPB SOCKET interface

FPGA IBM uses the CoreConnect Xilinx bus as the FPGA chip interconnects bus.
CoreConnect IBM bus is designed and developed by IBM for the SOC chip bus
communication chain; it can make a number of IP chips to connect with each other to
become a complete new chip. IBM coreconnect bus architecture consists of processor
local bus processor local bus (PLB), on-chip peripheral bus on-chip peripheral bus
(OPB), bus bridge, arbiter, and device control register (device control register DCR) bus.
The design of wearable computer SOPC prototype system, using the above three kinds of
bus, the PLB bus for connecting processor, Bram controller and user-defined VGA core
and DDR2 Memory Controller; OPB bus for connecting serial IP core, user custom LCD
nucleus, system ace controller and related peripheral controller; DCR bus for connection
control registers of the bus macros and reconfigurable module. CoreConnect IBM chip
bus technology to make integration easier, but also in the design of the processor core,
peripheral IP core can be reused in order to achieve a higher overall system performance.
The entire wearable computer hardware system bus connection can be set in the EDK
development environment, as shown below:
452 RISTI, N.º E5, 07/2016

Figure 6 – EDK wearable computer SOPC bus connection
4. Software design of wearable computer

4.1. Linux system transplantation
Wearable computer can be used in complex field environment, with more functional
requirements. In order to achieve the multi tasking requirements of wearable computers,
as well as the various features of the day after the expansion, we need to transplant
the multi tasking Linux system. This will lay the foundation for further research in the
laboratory. The following focuses on the ML410 development platform for the Linux
operating system migration process. Before Linux transplantation, need to establish a
cross compiler environment, so as to compile the kernel in the back. The so-called cross
compiler is compiled on the host computer to generate the target machine PowerPC405
processor executable binary bit stream file. First need to build a Linux virtual machine
environment on the PC machine, otherwise the transplant work can not be carried out.
After the establishment of a virtual machine, but also need to set up a shared folder
windows and Linux, this paper is set to D:/share environment windows. In the virtual
machine environment is /mnt/hgfs/share.
Before building the cross compile environment, you need to prepare the following
documents: crosstool-0.43.tar.gz, gcc3.4.4.tar.gz, glibc2.3.5.tar.gz, binutils-2.15.tar.gz,
glibc-linuxthreads-2.3.5.tar.gz. These are indispensable documents for transplantation.
After the above documents are ready, you can enter the virtual machine linux environment
to operate, the specific operation of the command as follows:
#su root
#mkdir /mnt/linux/crosstool
#chown username /mnt/linux/crosstool/
#exit
#cd /mnt/linux/crosstool/
#tar zxvf crosstool-0.43.tar.gz
#cd crosstool-0.43
#vim demo-powerpc-405.sh
RISTI, N.º E5, 07/2016 453

BSP EDK in the Linux options, in the menu Software open Platform Settings Software
window can be directly selected:
Figure 7 – Selection of Linux system in EDK software setting interface

The parameters needed to be set in the Library OS interface are shown in the
following figure:
Figure 8 – OS and Library settings
After specifying the parameter settings, select the Libraris and BSPs Generate in the
menu Software to generate the required BSP. After a period of compilation, the EDK
project directory will appear in two directories: arch and drivers. The arch is associated
with the plate structure of the BSP, the drivers are peripheral driver code, the directory
Arch\ppc\platforms\xilinx_ocp OCP is an important document Xparameters_ml300
Xparameters_ml300. It records the hardware parameters of the operating system, is the
main components of the address information and the clock frequency. We can modify the
ML300 board Linux code to generate the code to adapt to our ML410 development board.
454 RISTI, N.º E5, 07/2016

4.2. Software implementation of dynamic reconfiguration

In the dynamic reconfiguration system of this paper, the use of ICAP port is the key to the
design of the system software. Only the correct part of the configuration of the bit files
through the ICAP port to write the FPGA device, to be able to reconstruct the reconfigurable
resources. Some of the configuration files are easily transferred to the FPGA device through
the ICAP port configuration, so as to achieve the operation of self reconfiguration. The
whole dynamic reconfiguration of software development can be carried out in the EDK
development environment, and do not need to go in the Linux environment. Because
the EDK tool is very powerful, it can be used in SDK application software development,
and compiled into the target board software configuration file. First of all, through the
API function XHwIcap_StorageBufferWrite XHwIcap_StorageBufferWrite the actual
configuration data transfer to HWICAP Bram; and when after the full of HWICAP
Bram, through the API function XHwIcap_DeviceWrite XHwIcap_DeviceWrite the
configuration data write ICAP port. ICAP initialization code as shown below:
Figure 9 – ICAP initialization code

XHwIcap_Initialize () for the ICAP initialization function, the function of the three
parameters, the first one is a pointer to the function of the HwIcap instance, second for
the detection of the device ID, the third is the use of the ID FPGA. The function to load
the configuration data stream from the:
XHwIcap_CF2Icap (*hwicap XHwIcap, filename Xuint8*)
The main function of the system through the hard PowPC control dynamic reconfiguration
of the operating process, you first need to the control signal of the logic circuit for reading,
then according to the different control signals into different reconfiguration strategies,
mainly the following code:
XIo_Out32(XPAR_OPB_GPIO_1_BASEADDR,0x00000000);
while (1) {
Reg32Value = XIo_In32(XPAR_OPB_GPIO_0_BASEADDR);
switch (Reg32Value) {
case ‘1’:
lcd++;
xil_printf(“Performing reconfiguration for lcd\r\n”);
xil_printf(“-----------------------------------------------------------\r\n”);
xil_printf(“onload lcd module for display information.\r\n”);
XHwIcap_CF2Icap(&MyIcap, “ reconfig_lcd.bit”);
RISTI, N.º E5, 07/2016 455

delay(2000000);
break;
case ‘2’:
vga++;
xil_printf(“Performing reconfiguration for vga\r\n”);
xil_printf(“-----------------------------------------------------------\r\n”);
xil_printf(“Your press is 2,vga will slidshow the picture...\r\n”);
XHwIcap_CF2Icap(&MyIcap, “ reconfig_vga.bit”);
delay(2000000);
break;
case ‘3’:
all++;
xil_printf(“Performing reconfiguration for lcd and vga\r\n”);
xil_printf(“-----------------------------------------------------------\r\n”);
xil_printf(“Your press is 3,lcd will display char information and vga will slidshow the
picture...\r\n”);
XHwIcap_CF2Icap(&MyIcap, “ reconfig_all.bit”);
delay(2000000);
break;
default:
xil_printf(“configuration done! \r\n\r\n”);
xil_printf(“Press 1 for lcd display char information..\r\n\r\n”);
xil_printf(“Press 2 for vga will slideshow picture..\r\n\r\n”);
xil_printf(“Press 3 for lcd display and vga slideshow..\r\n\r\n”);
xil_printf(“Press r to restart system.. \r\n\r\n”);
break;
}
}
Function of the code is PowerPC processor reads the control signal information, then
Reg32Value values of the corresponding selection of reconstruction strategy according
to the, loading the corresponding reconstruction module to realize different function
applications.
4.3. Implementation of dynamic reconfiguration module

In this paper, the design of wearable computer system is carried out by using the
software and hardware coordination method. Build a hardware platform and software
development, the need for Xilinx powerful design and development tools support. In
this paper, the design and use of Xilinx provided by ISE9.1, EDK9.1, PlanAhead9.2.7
development tools to complete the design of the entire system. First, the wearable
computer system design is divided into several parts, and each part of the system is
divided into several modules. Which static module in during the running of the system
does not change, corresponding to the LED module, UART module, systemace die block
and DDR 2 module; and dynamic function module will change along with changes in
456 RISTI, N.º E5, 07/2016

the functional requirements, in the running process of the system will re allocation,
corresponds to the LCD and VGA module to display the character of the system
information and picture information. In planahead new projects, comprehensive import
netlist file; then divide the regional restructuring, the zoning of the overall design in
physics; the bus macro cloth in reconstruction of the regional CLB bar, position close
to the edge of the part of the reconstruction of the regional and across dynamic and
static circuits. PlanAhead will automatically generate user defined files for each dynamic
module. The static circuit is mapped to a single reconstruction region using the AREA_
GROUP constraint, which is not conducive to the stability of the system. The position
of the bus macro is precisely located between the reconfigurable module and the static
module, wherein, the four gate three state buffer is in one module, and the other four
gate three state buffers are in the other module.
Figure 10 – Addera_with_LEFT_a1 bus macro

All of the above constraints on the composition of the top-level design of the constraint
file top.ucf. By busmacro_xc4v_l2r_sync, _narrow.nmc top.ucf and busmacro_xc4v_
r2l_sync_narrow.nmc three files can be top-level design of the pre compiled. After
placing the good bus macro (BusMacro) in planahead environment also placed global
clock DCM, in planahead engineering input netlist, and properly positioned the bus
macro and global clock DCM needs after reconstruction of part of the design rules (DRC)
inspection, to ensure that the design is able to achieve a correct. Xilinx company to design
a set of verification rules for partial reconfiguration process, to ensure that the bus macro
can be correctly and placement, this set of rules also includes some other rules of the
reconstruction. If the design rules check the error, then part of the reconstruction process
cannot be completed, it cannot generate partial reconstruction of the bit stream file.
5. Conclusion
In this paper, based on the application requirements of wearable computer, the
dynamic reconfiguration module based on SOPC is realized. First in the modular
design methodology based on the determined the overall architecture of the dynamic
reconfigurable system, and the division of the module is fixed and reconfigurable module.
Then the integrated development environment ise and embedded development kit EDK
RISTI, N.º E5, 07/2016 457

and Xilinx specifically for the reconstruction of part of the design tools plan ahead of
wearable computer hardware and software development, and dynamic reconfigurable
module design; finally in Xilinx virtex-4 ml410 development platform to download
verification. Through the test and analysis of the performance of the configuration,
the correctness and efficiency of the proposed design method is verified. The dynamic
reconfiguration technology and SOPC design is applied to the wearable computer, which
has the advantage of the wearable computer based on COTS technology.
References
Asim, S., Daniel, S. (2012). Application Design For Wearable and Context-Aware
Computer. Pervasive Computing, IEEE, 20 (4), 20–29.
Becker, F. (2002). Configurable Systems-on-Chip: Commercial and Academic
Approaches.9th International Conference on Electronics, Circuits and Systems,
2(2), 809–812.
Fang, M., Chen, Z. (2014). User defined logic in SOPC design. Computer engineering,
30(17), 42–44.
Lysaght, P., Blodget, J. (2006). Enhanced Architectures, Design Methodologies and CAD
Tools for Dynamic Reconfiguration of Xilinx FPGAs. International Conference on
Field Programmable Logic and Applications, 1–6.
Zhu, M., Wen, Y. (2014). Research on FPGA dynamic reconfigurable digital circuit fault
tolerant system. Journal of Southeast University, 30 (4), 138–142.
458 RISTI, N.º E5, 07/2016

Design and Implementation of Communication

System Based on Android Platform
Mingming Chen 1,2,3, Hong Tang1,2, Yuzhi Chen1,2, Ning Wang 1,2,4*
* wning97@163.com
1
Department of Information and Mechatronics Engineering, Xiamen Huaxia University, Xiamen, China
2
Fujian Province Engineering Research Center on New Generation of Information and Communication
Technology and Wisdom Education
3
University of Illinois at Springfield, United States
4
Department of Automation, Xiamen University, Xiamen, China
Pages: 459–470
Abstract: This paper designs and implements a real time communication system
based on XMPP protocol, which is based on the open source server software
Openfire and Android platform. The system implements the basic functions of
message receiving and dispatching, meeting discussion group, file transfer and
user management. Finally we construct a perfect instant communication system
model, build real-time communication system for individuals or enterprises. The
communication system can reduce production and operation costs, and solve the
internal data transmission security. From the test results, the instant messaging
system can achieve user registration, user login, friend groups, data settings, as well
as mobile phone information interaction and other functions.
Keywords: Communication system, Android platform, XMPP protocol, Jive
Software
1. Introduction
In the 21st century, the instant messaging system as a kind of emerging technology
obtained the rapid development, this system as the mobile phone network, between
the different users in real time with other online users of information interactive
communication (Dongdong, 2013; Delgado, 2015). Its prominent advantage is fast and
convenient. Therefore, since its inception is favor of service industry and commercial
field. Most of traditional instant messaging application is implemented on the PC side,
but with the development of wireless from traditional Internet era into the era of mobile
Internet, there is much popular instant messaging software, such as AOL, MSN, QQ, and
Skype. Whether for individuals or businesses, the demand for instant messaging software
has been increasing. But these instant communication systems to the communication to
facilitate the same time, there have been new problems (Francisco, 2013; Jiménez, 2015).
First of all, there are a lot of data in the enterprise network transmission has a serious
security risk. Secondly, IM which according to their different protocols, resulting in these
RISTI, N.º E5, 07/2016 459

Design and Implementation of Communication System Based on Android Platform
systems (such as QQ, MSN, Ali wangwang, popo, Gtalk, etc.) The XMPP protocols appear
to solve this problem, XMPP Extensible messaging and presence protocol is a highly free,
highly open protocol based on extensible markup language (XML). Any software that
follows the XMPP protocol can communicate with each other (Zheran,2014; Ilker, 2015;
Rattanathip, 2015). XMPP has a good extensible architecture, can be easily integrated
into other applications, in the enterprise collaborative office, ERP, CRM and other large
systems have a very good development prospects. At present, including the Google
Company, Apple Corp, including the software giant also supports the XMPP agreement;
promote the promotion and popularization of the XMPP agreement.
This paper designs and implements a real time communication system based on XMPP
protocol, which is based on the open source server software Openfire and Android.
The system implements the basic functions of message receiving and dispatching,
meeting discussion group, and file transfer and user management. This paper focuses
on the research of XMPP protocol and Android inters process communication and
other technologies. Final development realized a set of relatively complete instant
communication system model, for individuals and businesses to build their own
instant messaging system, to reduce production and operating costs, solve internal data
transmission security, implementation and other communication system interoperability
issues put forward feasible solutions. After the completion of the system development,
author of the test run, test results from the point of view, the instant messaging system
can smoothly realize user registration, user login, grouping of friends, data set and
mobile phone information interaction function.
2. The Android platform and XMPP

2.1. Android platform
With the Android system mobile phone market share continues to rise, Android system
has attracted a large number of program developers to develop programs to expand the
function of Android mobile phone. Compared with other mobile phone operating systems,
Android has the following advantages: openness, equality, unbounded, convenience,
the richness of hardware. At present, the Android electronics market has more than
70000 applications, the total number of applications submitted by the developer more
than one hundred thousand. Android core intelligent communication service module
can be said to rely solely on the Linux 2.6 operating system kernel. Therefore, it has
Figure 1 – Hierarchical structure of Android system
460 RISTI, N.º E5, 07/2016

great similarity, such as efficient memory management, high safety, reasonable process
management, sound driven model and network protocol stack. Therefore, the kernel of
the Linux operating system also becomes a layer of abstraction between the software and
the hardware of the mobile phone. As shown in Figure 1, from a static point of view, we
can image the Android system is divided into four relatively independent layers”. But
from the point of view of the mobile phone operating system, Android system should
correspond to the various applications of the Linux system, an independent process
Libraries is the underlying core library, these programs are used by different components
in the FRAMEWORK APPLICATION layer, FRAMEWORK APPLICATION layer is an
object oriented package for these libraries. These libraries through the FRAMEWORK
APPLICATION layer for the implementation of the development of the technical staff
to provide a variety of required services. Usually contains the following core libraries:
•• Libc: system of C type library and the type library has certain directional
demand, is dedicated to the development of Linux system, can meet the
requirement of the embedded special demand, second is a development from
the BSD architecture evolution and the is a to C language as the language
foundation meet the view function library.
•• Media framework: multimedia type library and the type library mainly based
on Open CORE Packet Video language; it supports commonly used network and
mobile phone player of video and audio files such as MPEG4, MP3, AAC, BMP,
JPEG file type, and so on.
•• Manager Surface: This program library can not only achieve the 3D effect
and 2D effect of technology integration, but also to achieve the success of the
front-end display subsystem of the video and audio to play a prominent effect of
directional management.
•• LibWebCore: it is a browser database engine based on Web Web access mode,
which provides technical support for open mobile web pages to ensure the speed
and effect of the opening.
•• SGL: the bottom of the 2D class graphics display engine, in order to save the
resources of the way to open a simple 2D class graphics.
•• 3D libraries: mainly using the core data of OpenGL ES1.5 APIs to realize 3D
effect. The library operation easy to optimize the 3D software to realize the 3D
hardware acceleration, and better play software and hardware.
•• SQLite: This is a lightweight relational data engine, which can be easily invoked
by all applications.
Android applications are written in the Java programming language, they run in the
Dalvik virtual machine. Android application is by four components with each other,
mutual support after combination, these four components respectively for users to adjust
the parameters of the activity is responsible for the presentation layer, is responsible
for customer service service component, responsible for data acquisition and storage of
the content provider component, responsible for call service an intent receiver. Activity
includes: onCreate (), onStart (), onResume (), onPause (), onStop (), onDestory (). In
addition to this one and onRestart (), as shown in figure 2.
RISTI, N.º E5, 07/2016 461

Figure 2 – Android Activity life cycle
2.2. XMPP protocol

Extensible messaging and presence protocol (XMPP) is successful play the advantages
of XML structured data, the technical agreement appears success will attend and
context sensitive information directly embedded to the transformation of XML data,
it is the transformation of the data to be able to be selective and efficient transfer to
the definition of resources. Because XML program can easily through a variety of
firewalls, this advantage makes XMPP based technology for instant messaging services
and a variety of applications such as data from the mobile phone or Internet firewall.
So XMPP protocol can not only request for processing real-time news information,
also can response procedures need a variety of service, this kind of protocol can also
provide interoperability channels between different types of service of the instant
messaging system. The XMPP protocol RFC documents, general system platform and
no provisions of the developers or users must use the network topology, however,
many mainstream instant messaging providers usually use client / server architecture
to achieve. Client users usually use TCP connection to access the XMPP protocol
support server, and in different servers usually use TCP connection to communicate.
However, different from the common C/S architecture, the framework does not only
use an independent central server. The schematic diagram of this abstract architecture
is shown in Figure 3.
462 RISTI, N.º E5, 07/2016

Figure 3 – XMPP system structure

It is well known that the communication channel between the server and the client must
be authenticated before the exchange of information and the communication channel
for data communication and sharing. Prior to the establishment of communication
channels, in order to guarantee the data flow is a safe and effective transmission,
between two communication hardware platform must first negotiate TLS authentication
protocol, when TLS protocol negotiation is successful again SASL negotiation, SASL
agreement after the negotiation is successful to send signal and link session, get a list
of friends, resource binding. Figure 2-7 shows the structure of the security layer of the
instant messaging system under the XMPP protocol. The TCP protocol is an object -
oriented service layer, which can be accessed and accepted by all users. TLS protocols
Figure 4 – XMPP communication security mechanism
RISTI, N.º E5, 07/2016 463

are typically provided by the server’s operating system layer, which is used to ensure the
integrity and confidentiality of data communication between the two instant messaging
applications. SASL is usually provided by the application layer, the XMPP protocol is the
instant messaging system application itself.
3. Communication system design

3.1. Integrated architecture design
The system architecture design pattern is the client / server B/S mode, mainly to consider
the smart phone is its main users. Therefore in the process of the design of the user client
this paper focuses on the use of this model is mainly to take into account it is easier to
achieve information exchange and interaction. On the server side, this paper mainly uses
the open source server Openfire architecture and MySQL database model based on Java
development language. The request of the instant communication system sent by the
intelligent mobile phone user is processed by the request of the authentication request,
the data connection request, the information exchange request, the communication
protocol security confirmation and so on are processed by the Openfire server. The
architecture of this system is shown in figure 5.
Figure 5 – System architecture diagram
3.2. Server software design

In the system development and maintenance process, Jive software architecture is mainly
used for maintenance of server system. In addition, the openfire using open-source XMPP
security service agreement, to achieve the technical visits the server software system.
Opemfire deployment and management is very convenient, and has a good security
and scalability. It is based on the advantages of Openfire in the use of convenience,
scalability, and strong versatility, after comparison, this paper developed the instant
messaging system software decided to use the Openfire software architecture. In today’s
software development field, a mainstream guiding ideology is to avoid the disorder and
waste of software development, and to develop reusable software as far as possible from
the technical means. After nearly five years of development, in the instant messaging system
using the XMPP security protocol developers have been able to have a relatively complete,
open source, strong reusability of the foundation library support. Therefore, this system
464 RISTI, N.º E5, 07/2016

also uses the current development is relatively mature, the model is relatively fixed open
source library, based on this to develop real-time communication system required for the
various functions. Figure 6 shows the contrast before and after the Openfire server to add
the open source plug-ins. Table 1 lists the functions of these open source plug-ins.
Figure 6 – Openfire plug-in interface comparison
Open source plug-in describe

Asterisk-IM Openfire Plugin Asterisk and Openfire instant messaging integration
Broadcast Broadcast message to user
Client Control Control client connection
Content Filter Scan message packet
Email Listener Mail listener
Fastpath Service Managing chat requests in a queue
Monitoring Service Show statistical information about dialog and service
MotD Allows administrators to send messages
Packet Filter Packet filter
Presence Service Public user online status information via HTTP
Registration Perform various operations
Search User search
Table 1 – Openfire open source plugin

Under normal circumstances, openfire server software development technology standard
has good versatility, the popular database schema can be their realization technology of
match and fusion, and data reduction can meet the requirements of data storage mode
and follow-up technical support. For example, instant communication system commonly
used in the database have Sybase and Microsoft SQL server, mysql, Oracle, SQL server
and other, and these data as the back-end database server, the database technology
is relatively mature, capable of providing stable service. After comparison, we found
that the data support database of the instant communication system developed in this
paper will also use the Mysql database, because it can meet the requirements of software
development, the core of the system database table as shown in figure 7.
RISTI, N.º E5, 07/2016 465

Figure 7 – System database core relationship
3.3. The Android client design

Android Google operating system, the system user side of the various applications used
by the division is the way through the Java code or XML file to design and distribution.
XML file distribution is the way the development of the current development of the
popular way, but also the most widely used existing users. The Android operating system
for the user to provide a variety of layout of controls, such as currently used more a
relativelayout (relative layout), LineaLayout (linear layout), AbsoluteLayout (absolute
layout) and FrameLayou (single frame layout), tablelayout (table layout). Android
provides the application window controls, including Button, EditText, ProgressBar. In
the onCreate Activity () method using setContentView () to load the corresponding XML
layout file, the user can quickly display the custom of the software interface.
In the Linux system as the core technology of the Android system platform, the process
of inter - memory sharing becomes very important, and the Java program is precisely the
process of cross - process memory sharing cannot be achieved. Aidl first a language, is
used to describe the main interface of the communication system, compiler can be used
in the different language exchange process in the Aidl file some need to send information
into code, reuse defined between the different programming language interface to
achieve the purpose of two interposes communication process. In order to make it
easier for users to register, and for the development of procedures to provide ideas, the
author designed the instant communication system developed by registration process,
click the “register” button, the client automatically to the backend server to send a new
registration request, through the backstage after the confirmation, the system prompts
“register”, if failed to register, the system will prompt “failed to register”, and prompts
the user registration failure may be the cause. The key XML that the client sends to the
server registration request is as follows:
466 RISTI, N.º E5, 07/2016

User input user name and password, and then users click on the login button, the client
will send a login request to the server. After landing will display the corresponding main
interface, the failure will display the client tips.
Communication system of the client user interface information transfer or processing,

mainly including five: one is used to add new user “add contact module; the second is
to delete user user delete contact module; the third is used to change the user’s online
status of” change of state “module, is used for message processing message processing
module; the fifth is used to achieve the function expansion of the” other “operation
module. The client does not directly send the data directly to the backend server, it should
be processed through the XML data stream, and then use the interface conversion and
compiler, which will eventually transfer the data back to the server. The key XML that
the client sends to the server to add a contact request is as follows:
The key XML that the client sends to the server to delete a friend request is as follows:
The key XML that the client sends to the server to change the status request is as follows:
RISTI, N.º E5, 07/2016 467

4. Implementation of communication system

4.1. System deployment
Hardware environment: Server2003 Windows server two, respectively, equipped
with XMPP Openfire server, Mysql database. Containing the Android client cocom
of this system, test 3 PCS respectively equipped with spark client software. Software
environment: Android 2.2 platform, asmack-jse.jar, Eclipse, the server uses the
“Windows+Openfire+Mysql” model. After installing the Mysql database, and then
install the Openfire. Do not run after the installation of Openfire. Create an empty
database COCOM in Mysql, the Openfire installation directory under the database
directory to find the Openfire_mysql.sql file into the new library. The database you want
to configure the Openfire to use. Copy the lib to the mysql-connector-java-3.1.12-bin.jar
sub directory to install the Openfire directory, and then start the Openfire configuration
Openfire server environment.
Figure 8 – System test environment
4.2. Android client to achieve the main function

Smack is based on the Java language to write a XMPP implementation. Smack connection
server class XMPPConnection and package two constructors: XMPPConnection (string
ipAddr); through the constructor ipAddr to specified instant messaging platform
requires connecting to the XMPP server, and then enter the data call program calls
the function set real-time connect connection method. XMPPConnection (config
ConnectionConfiguration) in the connection function of the configuration parameter
ConnectionConfiguration for the connection class configuration, for the use of data
connection configuration parameters. For example, an instant messaging system
target server connection must first specify to carry out the port and IP address of the
connection to the target server, then according to the set procedure defines the security
settings (whether to use TLS encryption), followed by a definition program resource
name connection. After the user connection data of the instant communication system
is successful, the XMPP server requires the user to log in and enter the password
authentication to use the system’s relatively high level of transfer files and other services.
XMPPConnection class provides users with the login () method, used to authenticate
whether the client has been registered for use and password authentication to log on
468 RISTI, N.º E5, 07/2016

to the server. When the registered user login system, the client user can call the server
to provide the necessary operation of the expansion function. Smack communication
software to provide users with two mechanisms for processing functions, for real-
time XMPP information. The first is the PacketCollector (synchronous collector
mechanism), that is, the use of users in the XMPP connection to register a collector,
all of the XMPP information in accordance with the rules of all into this collector.
Another is PacketListener (the listener mechanism), is to use XMPP protocol provides
technical support is connected and register a listener, background server each received
a meet the set rules of information are handled by the listeners. Smack also provides
filter packet (filter), which allows the application to filter out irrelevant information,
the role of Smack as shown in figure 9. Due to the relatively strong smack database
functions, instant communication system development personnel does not have the
necessary basic care about relative underlying TCP send and receive data in detail, the
socket connection problems, and put major effort into ensuring top correctly received
and through the scientific design of transmission channels to ensure the smooth flow
of XML stream. Therefore, this system will use the Smack base library to achieve the
functional requirements of Android mobile client.
Figure 9 – The underlying communication support Android client
5. Conclusion
The communication system developed in this paper is based on the combination
of the real time communication application in fixed network and mobile network
terminal equipment, the XMPP protocol is used to realize the instant communication
system on the Android platform. In the development process of the system, this paper
on the theory of software engineering as a guide, the scrum agile development, with
sprint cycle on the progress of the project management, both in control of the project
or development cooperation are than the traditional way of development with greater
advantage. Also the project by the development mode of the software engineering, the
use of UML modeling, makes the analysis and design specifications has good readability
and intuitive, and the later expansion of secondary development has great guiding
role. XMPP protocol analysis, combined with the HTTP session model of the system is
established. For XMPP asynchronous transmission, using multiple threads to may cause
blocking operations to other threads to deal with, can make the wireless network of high
delay and low band broadband to the impact of reduced to the greatest extent. Instant
messaging has become the main technology of the on-line real-time communication of
voice and text, and its characteristics determine that it will play a more important role
in the future of mobile commerce, online collaboration and Internet applications. And
RISTI, N.º E5, 07/2016 469

with the rapid development of mobile communication technology, 3G network in China

put into commercial trial operation, provide more powerful support platform for mobile
instant messaging and real-time multimedia technology will transfer to the mobile
instant messaging application to. Integrated multimedia applications have a stronger
appeal, to provide users with more personalized services; will become an inevitable
trend in the future development of mobile instant messaging.
Acknowledgments
The work of this paper is supported by Education and Research Project of Fujian Province
(No.JA15877); Science and Technology Project of Xiamen (No.3502Z20163015).
References
Dongdong, L., Xianxiang, C. (2013). A Wireless Health Monitoring System based on
Android Operating System. IERI Procedia, 4, 208–215.
Delgado, A., Velthuis, M. (2015).Proposal for a continuous improvement IT governance
Francisco, G., Montoya, J. (2013). A monitoring system for intensive agriculture based on
mesh networks and the android system. Computers and Electronics in Agriculture,
99, 14–20.
Ilker, K., Senem, K. (2015). A cloud based and Android supported scalable home
automation system. Computers & Electrical Engineering, 43, 112–128.
Rattanathip, R. (2015). Development of Thai Rice Implantation Recommend System
Based on Android Operating System. Procedia - Social and Behavioral Sciences,
197, 1048–1052.
Zheran, F., Weili, H. (2014). Permission based Android security: Issues and
countermeasures. Computers & Security, 43, 205–218.
470 RISTI, N.º E5, 07/2016

Critérios Editoriais Criterios Editoriales
Critérios Editoriais Criterios Editoriales
A RISTI (Revista Ibérica de Sistemas e La RISTI (Revista Ibérica de Sistemas y

Tecnologias de Informação) é um periódico Tecnologías de la Información) es un periódico
científico, propriedade da AISTI (Associação científico, propiedad de la AISTI (Asociación
Ibérica de Sistemas e Tecnologias de Ibérica de Sistemas y Tecnologías de la
Informação), que foca a investigação e a Información), centrado en la investigación
aplicação prática inovadora no domínio dos y en la aplicación práctica innovadora en el
sistemas e tecnologias de informação. dominio de los sistemas y tecnologías de la
información.
O Conselho Editorial da RISTI incentiva
potenciais autores a submeterem artigos El Consejo Editorial de la RISTI incentiva
originais e inovadores para avaliação pelo autores potenciales a enviar sus artículos
Conselho Científico. originales e innovadores para evaluación por
el Consejo Científico.
A submissão de artigos para publicação na
RISTI deve realizar-se de acordo com as Lo envío de artículos para publicación en la
chamadas de artigos e as instruções e normas RISTI debe hacerse de conformidad con las
disponibilizadas no sítio Web da revista llamadas de los artículos y las instrucciones y
(http://www.risti.xyz). normas establecidas en el sitio Web de la revista
(http://www.risti.xyz).
Todos os artigos submetidos são avaliados
por um conjunto de membros do Conselho Todos los trabajos enviados son evaluados por
Científico, não inferior a três elementos. un número de miembros del Consejo Científico
de no menos de tres elementos.
Em cada número da revista são publicados
entre cinco a oito dos melhores artigos En cada número de la revista se publican cinco
submetidos. a ocho de los mejores artículos enviados.
RISTI, N.º E5, 07/2016 471

Formulário inscrição associado da AISTI Formulario inscripción asociado de la AISTI
Os associados da AISTI recebem a RISTI Los asociados de la AISTI reciben la

gratuitamente, por correio postal. RISTI por correo, sin costo alguno.
Torne-se associado da AISTI. Preencha Hazte miembro de la AISTI. Rellena
o formulário abaixo e envie-o para o el siguiente formulario y remítelo al
e-mail aistic@gmail.com e-mail aistic@gmail.com
Formulário de Associado / Formulario de Asociado
Nome/Nombre:
Instituição/Institución:
Departamento:
Morada/Dirección:
Código Postal: Localidade/Localidad:
País:
Telefone/Teléfono:
E-mail: Web:
Tipo de Associado e valor da anuidade:
Individual - 25€
Instituição de Ensino ou I&D/Institución de Educación o I&D - 250€
Outro (Empresa, etc.) - 500€
NIF/CIF:
Data/Fecha: / / Assinatura/Firma:
472 RISTI, N.º E5, 07/2016

Revista lbérica de Sistemas e Tecnologias de Informação
Revista lbérica de Sistemas y Tecnologías de Información ©AISTI 2016 hhttp://www.risti.xyz
RISTI, N.º E5, 07/2016 473

Ristie 5

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ristie 5

Uploaded by

Copyright:

Available Formats

ISSN: 1646-9895

Revista Ibérica de Sistemas e Tecnologias de Informação

©AISTI 2016 http://www.aisti.eu Nº E5

Propriedade e Publicação / Ownership and Publication

RISTI, N.º E5, 07/2016 i

Coordenadores da Edição / Issue Coordinators

Conselho Editorial / Editorial Board

Conselho Científico / Scientific Board

ii RISTI, N.º E5, 07/2016

RISTI, N.º E5, 07/2016 iii

iv RISTI, N.º E5, 07/2016

RISTI, N.º E5, 07/2016 v

vi RISTI, N.º E5, 07/2016

RISTI, N.º E5, 07/2016 vii

Information Technology Drives Innovation in Service

Renmin University of China, Beijing, China

Technological innovation presents a diversified development trend, with the integration

RISTI, N.º E5, 07/2016 ix

x RISTI, N.º E5, 07/2016

RISTI, N.º E5, 07/2016 xi

xii RISTI, N.º E5, 07/2016

RISTI, N.º E5, 07/2016 xiii

Research on Influence Factors of New Rural Sports

RISTI, N.º E5, 07/2016 1

2. Data mining and the rural information network

2 RISTI, N.º E5, 07/2016

Figure 1 – The data mining process and influence

Figure 2 – Itemsets transcendental principle

RISTI, N.º E5, 07/2016 3

Figure 3 – Support-Based Pruning

2.2. The construction of rural information network

4 RISTI, N.º E5, 07/2016

2. Rural network culture constructions are an important platform to improve the

3. The research object and methods

3.2. Main research methods

RISTI, N.º E5, 07/2016 5

variable indicators measuring scale and in formal empirical research by larger

Initial Eigenvalues Rotation Sums of Squared

Table 1 – Total Variance Explained

4.2. Classification of factors

6 RISTI, N.º E5, 07/2016

Table 2 – Component Matrix

RISTI, N.º E5, 07/2016 7

4.3. New development strategy of sports culture

indicators Mean value T value P value ranking

Table 3 – The Jiangxi government behavior index(n=223)

8 RISTI, N.º E5, 07/2016

indicators Mean value T value P value ranking

Table 4 – Jiangxi rural economic index important degree (n=223)

RISTI, N.º E5, 07/2016 9

indicators Mean value T value P value ranking

Table 5 – Jiangxi rural sports culture life index (n=223)

indicators Mean value T value P value ranking

Table 6 – Jiangxi rural environment index (n=223)

10 RISTI, N.º E5, 07/2016

indicators Mean value T value P value ranking

Table 7 – Jiangxi rural social influence index(n=223)

RISTI, N.º E5, 07/2016 11

12 RISTI, N.º E5, 07/2016

Influence of Corporate Equity Incentive and CSR on

Ming Li 1, Zhenxiang Chen 1,2,*

Abstract: According to the stakeholder theory, employees are important

RISTI, N.º E5, 07/2016 13