Welcome to the InfoSci Platform
Reference Hub2
Combining Trust Propagation and Topic-Level User Interest Expansion in Recommender Systems

Combining Trust Propagation and Topic-Level User Interest Expansion in Recommender Systems

Zukun Yu, William Wei Song, Xiaolin Zheng, Deren Chen
Copyright: © 2016 |Volume: 13 |Issue: 2 |Pages: 19
ISSN: 1545-7362|EISSN: 1546-5004|EISBN13: 9781466689053|DOI: 10.4018/IJWSR.2016040101
Cite Article Cite Article

MLA

Yu, Zukun, et al. "Combining Trust Propagation and Topic-Level User Interest Expansion in Recommender Systems." IJWSR vol.13, no.2 2016: pp.1-19. http://doi.org/10.4018/IJWSR.2016040101

APA

Yu, Z., Song, W. W., Zheng, X., & Chen, D. (2016). Combining Trust Propagation and Topic-Level User Interest Expansion in Recommender Systems. International Journal of Web Services Research (IJWSR), 13(2), 1-19. http://doi.org/10.4018/IJWSR.2016040101

Chicago

Yu, Zukun, et al. "Combining Trust Propagation and Topic-Level User Interest Expansion in Recommender Systems," International Journal of Web Services Research (IJWSR) 13, no.2: 1-19. http://doi.org/10.4018/IJWSR.2016040101

Export Reference

Mendeley
Favorite Full-Issue Download

Abstract

With the development of E-commerce and Internet, items are becoming more and more, which brings a so called information overload problem that it is hard for users to find the items they would be interested in. Recommender systems emerge to response to this problem through discovering user interest based on their rating information automatically. But the rating information is usually sparse compared to all the possible ratings between users and items. Therefore, it is hard to find out user interest, which is the most important part in recommender systems. In this paper, we propose a recommendation method TT-Rec that employs trust propagation and topic-level user interest expansion to predict user interest. TT-Rec uses a reputation-based method to weight users' influence on other users when propagating trust. TT-Rec also considers discovering user interest by expanding user interest in topic level. In the evaluation, we use three metrics MAE, Coverage and F1 to evaluate TT-Rec through comparative experiments. The experiment results show that TT-Rec recommendation method has a good performance.

Request Access

You do not own this content. Please login to recommend this title to your institution's librarian or purchase it from the IGI Global bookstore.