You are on page 1of 36

Under the Guidance of:

Dr. Bharti Joshi

Presented by:
Suhasini Parvatikar

INDEX
Recommender system
Content Based Filtering
Collaborating Filtering
User based CF
Item based CF

Existing System
Proposed System
9/2/2015

Recommender Systems
A recommender system (RS) helps people that have

not sufficient personal experience or competence to


evaluate the number of alternatives offered by a Web
site.
In their simplest form Recommender system recommend to

their users lists of items


Provide consumers with information to

help them decide which items to purchase


3

9/2/2015

Mainly there are 2 approaches of Recommender System


Content Based filtering
Collaborative Filtering

9/2/2015

Content Based Filtering


In content-based recommendations the system tries to

recommend items that matches the User Profile.


The Profile is based on items user has liked in the past or

explicit interests that he defines.


A content-based recommender system matches the profile of

the item to the user profile to decide on its relevancy to the


user

9/2/2015

Simple Example

Fig_name: Content Based Filtering

9/2/2015

Drawbacks:
Content based filtering is also having some limitations like

finding the quality of the content.


For example: Content based filtering cannot differentiate

between good article and bad article if both of them are


using same terminology
User cold-start problem
What if users interests change?

9/2/2015

What is Collaborative
Filtering?
Community of users
To predict a users opinion, use the opinions of others
Collaborative filtering uses 2 approaches to predict and

recommend
k-nearest neighbor
association rules based prediction

9/2/2015

Examples of Collaborative
Filtering

Fig_name: Collaborative Filtering


9/2/2015

Nearest Neighbor Classifiers


Basic idea:
If it walks like a duck, quacks like a duck, then its probably a duck

Compute
Distance

Training
Records

Test
Record

Choose k of the
nearest records

10

Association Rule Mining


Association rules obviously can be used for recommendation.
Each transaction for association rule mining is the set of items

bought by a particular user.


We can find item association rules, e.g.,

buy_X, buy_Y -> buy_Z


Rank items based on measures such as confidence, etc.

9/2/2015

11

Types of Collaborative
Filtering
User-based collaborative filtering
Item-based collaborative filtering

9/2/2015

12

User-based Collaborative
Filtering
To predict a users opinion for an item , use the opinion of similar users.

A user-based collaborative filtering method consists of two primary phases:


the neighborhood formation phase
the recommendation phase.

9/2/2015

Fig_name: User Based Approach


13

9/2/2015

14

Example:

9/2/2015

15

Similarity between users


How similar are users 1 and 2?

9/2/2015

16

Similarity Calculation
Similarity is calculated by Pearson correlation coefficient

9/2/2015

17

Recommendation Phase
Use the following formula to compute the rating prediction of item i for target user u

where V is the set of k similar users, rv,i is the rating of user v given to item i.
For each item:

p (u, i ) r

vV

Calculate difference in the users


u ratings

Take the average of this difference over the items

9/2/2015

sim(u, v ) (rv ,i r v )

vV

sim(u, v )

18

User Cold-Start problem: not enough known about new user to

decide who is similar


Sparsity: when recommending from a large item set, users will

have rated only some of the items

Problems with User-based CF


9/2/2015

19

Item Based CF
Similarity between items is decided by looking at how other users have rated them
Item based CF has 2 approaches:
Neighbourhood Formation Phase
Recommendation Phase

Fig_name: Item based approach


9/2/2015

20

Example

9/2/2015

21

Similarity between Items


How similar are

items 3 and 4?

9/2/2015

22

Similarity Calculations
Similarity can be calculated using Cosine similarity

9/2/2015

23

Recommendation Phase
After computing the similarity between items we select a set of k most similar items to the target item and generate a predicted value of user us rating

where J is the set of k similar items

p (u, i )

jJ

u, j

jJ

9/2/2015

sim(i, j )

sim(i, j )

24

Problems with Item-based CF


Item Cold-Start problem : Cannot predict which items are

similar till we have ratings for this item

9/2/2015

25

Existing System
Combines
Content Filtering
Collaborative Filtering(item based CF)
Association rule mining

But the drawback of this system is it doesnt solve cold start

problem i.e new user /new item which does not have any
ratings.
9/2/2015

26

So we propose Semantically Enhanced Item-based CF

Algorithm .

9/2/2015

27

Proposed System

9/2/2015

28

This project combines


Content Based filtering
Collaborative Filtering(Semantically Enhanced Item based CF)
Association Rule Mining

9/2/2015

29

Semantically Enhanced Item


Collaborative
Filtering
Basic Idea:
Extend item-based collaborative filtering to incorporate both similarity

based on ratings (or usage) as well as semantic similarity based on


domain knowledge

Semantic knowledge about items


Can be extracted automatically from the Web based on domain-specific

reference ontologies
Used in conjunction with user-item mappings to create a combined

similarity measure for item comparisons


30
9/2/2015

Semantically Enhanced Item


CF
An extension of the item-based algorithm
Use a combined similarity measure to compute item

similarities:

where,

SemSim is the similarity of items ip and iq based on semantic


features (e.g., keywords, attributes, etc.); and
RateSim is the similarity of items ip and iq based on user
ratings (as in the standard item-based CF)
is the semantic combination parameter:
= 1 only user ratings; no semantic similarity
= 0 only semantic features; no collaborative similarity

31

REFERENCES
[1] ANAND SHANKER TEWAR,ABHAY KUMAR,ASIM GOPAL BARMAN, Book
Recommendation System Based on Combine Features of Content Based Filtering,
Collaborative Filtering and Association Rule Mining,2014 IEEE.
[2] NIRAV M. KHETRA, SHRUTI B.YAGNIK A Collaborative Approach for Web
Personalized Recommendation System 2014 IJEDR Volume 2, Issue 4 ISSN:
2321-9939
[3] ROHINI NAIR, KAVITA KELKAR, Implementation of Item and Content

based Collaborative Filtering Techniques based on Ratings Average for


Recommender Systems, International Journal of Computer Applications (0975
8887) Volume 65 No.24, March 2013

9/2/2015

32

[4] A. KUMAR AND DR. P. THAMBIDURAI , Collaborative Web Recommendation Systems


-A Survey Approach, Global Journal of Computer Science and Technology Vol. 9 Issue 5 (Ver
2.0), January 2010
[5] FRANCESCO RICCI ,LIOR ROKACH, BRACHA SHAPIRA ,PAUL B.
KANTOR,Recommender system handbook,springer ,may 2010.
[6] GEDIMINAS ADOMAVICIUS ,ALEXANDER TUZHILIN, Toward the Next Generationof
Recommender Systems: A Survey of the State-of-the-Art and Possible Extensions IEEE
transactions on knowledge and data engineering, Vol. 17, no. 6, June 2005.
[7] EIRINAKI, M. AND VAZIRGIANNIS, M.: Web Mining for Web Personalization. ACM
Trans. On Internet Technology, Vol. 3, 1 (Feb.,2003) .
[8] SARWAR, B., KARYPIS, G., KONSTAN, J., AND REIDL, J.2001 Item based collaborative
filtering recommendation algorithms. In Proceedings of the 10th International Conference on
World Wide Web (WWW01). ACM, New York.

9/2/2015

33

[9] SHARDANAND, U. AND MAES, P. 1995. Social information filtering: algorithms for
automating word of mouth. In Proceedings of the SIGCHI Conference on Human Factors in
Computing Systems (CHI95). ACM Press/Addison-Wesley Publishing Co., New York, NY.
[10] FOLTZ, P. W. AND DUMAIS, S. T. 1992. Personalized information delivery an analysis of
information filtering methods. Comm. ACM 35(12).
[11] MOHD ABDUL HAMEED, OMAR AL JADAAN,S. RAMACHANDRAM, Collaborative
Filtering Based Recommendation System: A survey, International Journal on Computer Science
and Engineering (IJCSE).
[12] ATISHA SACHAN, VINEET RICHARIYA, A Survey on Recommender Systems based on
Collaborative Filtering Technique, International Journal of Innovations in Engineering and
Technology[IJIET].
[13] PURVI DUBEY,PRAMOD S. NAIR, Recommendation System for Web Mining: A
Review International Journal of Computer Applications (0975 8887),Volume 109 No. 11,
January 2015
9/2/2015

34

.
[14] GREG LINDEN,BRENT SMITH, AND JEREMY YORK, Amazon.com
Recommendations Item-to-Item Collaborative Filtering, Jan 2003 IEEE Internet
Computing.
[15] www.google.com
[16] www.googleimages.com
[17] XIN JIN , BAMSHAD MOBASHER Using semantic similarity to enhance itembased collaborative filtering International Journal of Computer Application

9/2/2015

35

TH ANK YOU

9/2/2015

36

You might also like