Academic Journals Database
Disseminating quality controlled scientific knowledge

FARS: Fuzzy Ant based Recommender System for Web Users

ADD TO MY LIST
 
Author(s): Shiva Nadi | Mohammad H. Saraee | Mohammad Davarpanah Jazi | Ayoub Bagheri

Journal: International Journal of Computer Science Issues
ISSN 1694-0784

Volume: 8;
Issue: 1;
Start page: 203;
Date: 2011;
Original page

Keywords: Web personalization | Recommender Systems | Ant colony optimization | Fuzzy set | IJCSI

ABSTRACT
Recommender systems are useful tools which provide an adaptive web environment for web users. Nowadays, having a user friendly website is a big challenge in e-commerce technology. In this paper, applying the benefits of both collaborative and content based filtering techniques is proposed by presenting a fuzzy recommender system based on collaborative behavior of ants (FARS). FARS works in two phases: modeling and recommendation. First, user's behaviors are modeled offline and the results are used in second phase for online recommendation. Fuzzy techniques provide the possibility of capturing uncertainty among user interests and ant based algorithms provides us with optimal solutions. The performance of FARS is evaluated using log files of "Information and Communication Technology Center" of Isfahan municipality in Iran and compared with ant based recommender system (ARS). The results shown are promising and proved that integrating fuzzy Ant approach provides us with more functional and robust recommendations.
Affiliate Program     

Tango Jona
Tangokurs Rapperswil-Jona