1Building Recommendations by means of Classification Based on Fuzzy Association

Joel Pinho

Description

1Building Recommendations by means of Classification Based on Fuzzy Association Joel Pinho Lucas Supervisor: María N. Moreno García (MIDA Research Group) Collaboration: Anne Laurent, Maguelonne Teisseire (LIRMM, TaToo) PhD Program in Computer Science Department of Computer Science and Automatics University of Salamanca, Spain

• employ recommender


• based


• fuzzy association ?


• future works ?


• find products


• cba-fuzzy algorithm ?


• scenario ?

Sujets

• Association rule learning
• Amazon
• Laurent

Informations

• Publié par: pefav
• Nombre de lectures: 24
• Langue: English

Extrait

1

Building Recommendations by means of

Building Recommendations by means of

Classification Based on Fuzzy Association

Classification Based on Fuzzy Association

Joel Pinho Lucas

Supervisor: María N. Moreno García (MIDA Research Group)

Collaboration: Anne Laurent, Maguelonne Teisseire (LIRMM, TaToo)

PhD Program in Computer Science

PhD Program in Computer Science

Department of Computer Science and Automatics

Department of Computer Science and Automatics

University of Salamanca, Spain

2

Outline

Outline



Scenario

Scenario



Recommender Systems

Recommender Systems



Association Rules

Association Rules



Classification Based on Association

Classification Based on Association



Case Studies

Case Studies



Results

Results



Hybrid Recommendation Method

Hybrid Recommendation Method



Fuzzy Logic and Fuzzy Association

Fuzzy Logic and Fuzzy Association



CBA-Fuzzy Algorithm

CBA-Fuzzy Algorithm



Conclusions and Future Works

Conclusions and Future Works



References

References

3

Scenario

Scenario



It has been estimated that the amount of information in

the world doubles every 20 months (Breese, et. al.)



It is technologically feasible to store and publish huge

volumes of data with, more and more, lower costs



Nowadays it is harder to process and extract valuable

information from databases

4

Scenario

Scenario

E-commerce Systems:



The

“

information explosion

”

denoted nowadays is reflected by

loads of products available for sale



Difficulty in choosing and purchasing products



They need to personalize the presentation of their products



They employ recommender systems to help consumers in order

to find products to purchase (Thabtah et. al.)

5

Recommender Systems

Recommender Systems



Examples: MovieLens, Video Recommender, Amazon,

etc.



They can be divided into two main classes:



Content-based methods:

compare text documents

to user profiles



Collaborative filtering methods:

are based on

opinions of other users (

ratings

)



Memory based (User-based)



Model based (Item-based)