Integration and Interaction of Distributed Data Mining with Agent Technology

Amit Joshi#1, Bhavesh Joshi*2

Department of Computer Science & Engineering, Pacific University Pacific Hills, Udaipur (Rajasthan), India * Department of Computer Science, Advent Institute of Management Studies, Udaipur (Rajasthan), India
1

amitjoshiudr@gmail.com 2joshib05@gmail.com

Abstract In recent years, more and more researchers have been involved in research on both agent technology and distributed data mining. A clear disciplinary effort has been activated toward removing the boundary between them that is the interaction and integration between agent technology and distributed data mining. We refer this to agent mining as a new area. The marriage of agents and distributed data mining is driven by challenges faced by both communities, and the need of developing more advanced intelligence, information processing and systems. This paper presents an overall picture of agent mining from the perspective of positioning it as an emerging area. We summarize the main distributed data mining, driving forces, disciplinary framework, applications, and trends and directions, data mining-driven agents, and mutual issues in agent mining. Arguably, we draw the following conclusions: (1) agent mining emerges as a new area in the scientific family, (2) both agent technology and distributed data mining can greatly benefit from agent mining, (3) it is very promising to result in additional advancement in intelligent information processing and systems. However, as a new open area, there are many issues waiting for research and development from theoretical, technological and practical perspectives. Keywords Cryptography, Caesar Cipher, Plain Text, Cipher Text, Randomized.

All of the techniques has its own pros & cons. Out of the above three techniques the algorithm proposed by the paper is using Secret Key Cryptography (SKC) in which only one key used for encrypting and decrypting the information( See Fig- 1).

Fig 1- Conversion of Plain to Cipher text using secret key. PRE REQUISITES FOR CRYPTOGRAPHY In the process of encryption and decryption there are various parameters which have to be taken into consideration: Authentication: In the process of authentication process it is checked that weather the Sender of the message is authentic or not. Confidentiality: The secret of the encoding the message is kept constant and the message is not accessible to any third user. Integrity: Integrity means no data in the message has been altered in data/ message passage. Non-repudiation: It means the authentic information to prove that the sender has sent the message. Digitized proof for this pre requisite is very necessary.

INTRODUCTION There are many security implications and many applications, ranging from secure transmission of data, protecting passwords, On-line security, On-line banking and many more. One essential aspect for secure communications is cryptography which can be defined as the conversion of data into a ciphered code that can be deciphered and also could be safely send across a public or private network. Alone Cryptography it is not sufficient for secure transmission of information. So paper has devised an algorithm which is using a randomized technique to generate key along with Cryptographic Algorithm to encrypt and decrypt data [1]. CRYPTOGRAPHIC ALGORITHM There are several ways of classifying cryptographic algorithms. Since the paper is using an algorithm which is using random key generation, they will be categorized based on the number of keys that are employed for encryption and decryption, and further defined by their application and use. The three types of Techniques based on number of keys are: Secret Key Cryptography (SKC): Uses a single key for both encryption and decryption. Public Key Cryptography (PKC): Uses one key for encryption and another for decryption. Functions: Uses a mathematical Hash transformation to encrypt information [2].

OBJECTIVE OF PROPOSED ALGORITHM The main objective of the algorithm is to avoid passing of the secret key in the encrypted message from sender to the receiver and the algorithm aims at random generation of the secret key sender has used to encrypt the message at the receivers end. This totally changes a secret key technique in which key is passed by sender to the receiver. Random key generation technique helps in generation of the random secret at both the sender and receiver end. The proposed algorithm uses the global time of sending the message to generate a random secret key. This senders global time is retrieved through the gateway and random key is generated at receivers end.

CAESAR CIPHER TECHNIQUE In cryptography, a Caesar cipher, also known shift cipher, is one of the simplest and most widely known encryption techniques. It is a type of substitution cipher in which each letter in the plaintext is replaced by a letter some fixed number of positions down the alphabet. For example, with a shift of 3, A would be replaced by D, B would become E, and so on. The method is named after Julius Caesar, who used it to communicate with his generals [3]. The paper uses the technique and all versions generated by the different shift keys. We have taken all the 25 keys into consideration which could be used as shift keys for alphabets. The proposed algorithm has created a group of 25 keys for numbers and special characters as well ranging from 1 to 25 in order to maintain scalability. RANDOMIZED SECRET KEY GENERATION TECHNIQUE In order to generate a secret key various methods could be used. Random number generation technique could be one. But randomization technique cannot be applied directly because in case of secret key cryptography it is necessary that both the sender and receiver should know about the secret key. So if a secret key is randomly generated at the time of decryption of the message the retrieval of the secret key becomes a barrier. So in order to generate a secret key the proposed algorithm uses Global time which would be grasped at the gateway of the network. Registers are used to store the global time [4]. The secret key is calculated by adding the number of hours, minutes and seconds. If the total exceeds the maximum number of shift keys i.e. 25 the generated shift key is processed again. If the generated secret key is a even then it is divided by two else if the key is odd the generated secret key is multiplied by three and one is added to it. The process continues till the generated secret key becomes less than 25. After the whole processing the result we get is final randomly generated secret key.

HYPOTHESIS The following assumptions have been taken into consideration: The algorithm would be using 25 shift keys from 1 to 25. Each shift key would be shifting the characters according to its value. All the alphabets have been kept in one group and all numbers and special characters (except punctuations) have been kept in another group. The data groups have been shown in Table 1. In case of alphabets we can shift them to 25 locations according to 25 shift keys. Similarly numbers and special characters can also be shifted to 25 locations in same manner. The arrangements of the 26 alphabets and 10 numbers and special characters are grouped together as per the requirements of the proposed algorithm. Encryption and decryption both process will always be from beginning to the last. A blank space between the message words divides the message into different blocks. Each block is treated differently and a new shift key is generated for each block.
TABLE I DATA GROUPS FOR ENCRYPTION

CRYPTOGRAPHIC ALGORITHM & TECHNIQUE The basic steps of proposed algorithm are: The basic work is to find the Secret key. So at the gateway of the network the global time is grasped and stored in registers. Now to calculate secret key for the first block we just add number of hours, minutes and seconds. Thereafter a secret key is generated. If it exceeds 25 shift keys it is processed again and same process is done as mentioned in randomized technique. Now the first block of information is processed according to the secret key used as a shift key. As soon as a blank space is encountered, it signifies the end of first block. Now to encrypt second block a new shift key has to be generated. The new key is generated by the same technique as mentioned above that if the number is even, divide by 2 else multiply by 3 and add 1. If exceeds 25 the processing is done again till a number less than or equal to 25 is encountered.

The algorithm proceeds as such all the blocks are encrypted and the message ends. For decryption also same procedure is followed. Global time of message sending is recovered by the receiving network gateway and accordingly decryption takes place step by step.

The complete procedure of the algorithm or the steps of the algorithm are shown in Fig 2.

and basic operations the target is achieved. So a normal processor which can perform basic mathematical operations is enough for the algorithm. But for third party it is difficult is crack the generation of the secret which is actually the key requirement of the algorithm. And in order to decrypt the message all the possible combinations has to be tried which are 625 for above used message shown in Table 3. For decryption a large time and memory consumption is required for checking of each combination which is 625 for the presented data set.
TABLE IIIII VARIOUS COMBINATIONS FOR MESSAGE BITS

Fig 2 Block Diagram showing Encryption Process. IX. ALGORITHM IMPLEMENTATION The algorithm generates a 48 bit secret key randomly which gets converted to 16 bits secret key for the use in the algorithm. This 16 bit generated secret key gets applied on a block (maximum of 128 bits) one after the other till the whole message is encoded. The size of one block may vary from one to another block. Random secret key generated would be different for different blocks and would be generated according to the algorithm.. In order to implement the algorithm a text message is taken. UDAIPUR IS CITY OF LAKES is the text message to be passed to the receiver through the network after encrypting it. In order to generate the secret key global time of the message sending is recovered from mail server. At the instant the the recovered was 05:11:03 .Now 5+11+3 gives the initial secret key which comes to be 19. Since it is in range the encryption of first block starts till a space is encountered. So as per the shift key UDAIPUR gets converted to NWTBINK. Now for the next block the shift key is manipulated. Since it is odd it becomes (19*3) +1=58. Now it is divided by 2 and becomes 29. Again we calculate (29*3) +1=88 then 44 and then 22 which is the secret key for next block. Now IS is converted to EO. For the next block shift key is processed again and process continues till the message ends. Complete Decrypted keys are shown in Table 2. Decryption of the message also involves the same steps the only difference is that the message bits are shifted back forth according to the shift key.

The complexity of the algorithm is very low, since a fewer number of computation have to be performed for completion of secret key generation and the encryption of the data set and vice versa. The data set UDAIPUR IS CITY OF LAKES is encrypted to NWTBINK EO NETJ FW YNXRF using the algorithm. XI. COMPARISON WITH EXISTING ALGORITHMS

TABLE III ENCRYPTED MESSAGE

Table IV provides the comparison of the proposed algorithm with the existing algorithms and also provide key size and block used for effectively implementing the algorithm.
TABLE IVV COMPARISON WITH EXISTING ALGORITHMS

X. ALGORITHM COMPUTATIONS AND TESTING The algorithm was tested on Local Area Network (LAN). In the implementation of the message it was noted that 12 small computations were performed for generating the shift keys for different blocks which is the key task for the proposed algorithm. In addition to that 5 shift operations were performed on different blocks. So by using a total of 17 small

The algorithm can be easily implemented as there is no use of complex computation techniques and the algorithm relies on very basic computations techniques. The most important advantage of the algorithm is that no secret has to be passed along with message bits as it could be generated at the receivers side. The algorithm has used the existing Caesar Cipher technique manipulated the technique using random key generation. As the number of computations increases slightly the complexity increases but is balanced by the increasing number of combinations which can be randomly created. The number of combinations randomly increases to about 100 times the number of computations [5], [6]. A minimum number or no secret key has been to be passed to the algorithm in order to encrypt which is very low in comparison with other algorithms. The major advantage of the algorithm is the secret key is not passed with the encrypted message it could be easily calculated for the message decryption at the receivers end. XIII. CONCLUSION The paper has proposed an algorithm which can be easily and effectively used to send and transfer the data in any local area networks and private networks where security remains the prime implication for data passage. The paper also compares the designed security approach with the existing algorithms for security. The paper concludes that all results achieved so far are in alignment with the pre requisites and correct as per the proposed algorithm. REFERENCES On comparing with the existing algorithm it is being found that the secret key size is minimum for the randomize algorithm so a lesser time is consumed in order to encrypt a message with the algorithm and decreasing reliability is managed by randomly generation of key on both the ends. XII. PROS AND CONS OF PROPOSED ALGORITHM AND TECHNIQUE
Gary C. Kessler, An Overview of Cryptography , Handbook on Local Area Networks, September 1998. http://en.kioskea.net/contents/crypto/cleprivee.php3 dated 15 March 2011. Pieprzyk, Josef, Thomas Hardjono, Jennifer Seberry (2003). Fundamentals of Computer Security. Springer. pp. 6. ISBN 3-540-43101-2. Bamford, J. (1983). The Puzzle Palace: Inside the National Security Agency, America's most secret intelligence organization. New York: Penguin Books. Stallings, W. (2006). Cryptography and Network Security: Principles and Practice, 4th ed. Englewood Cliffs, NJ: Prentice Hall. Trappe, W., & Washington, L.C. (2006). Introduction to Cryptography with Codin Theory, 2nd ed. Upper Saddle River, NJ: Pearson Prentice Hall.