GroupLens: an open architecture for collaborative filtering of netnews
CSCW '94 Proceedings of the 1994 ACM conference on Computer supported cooperative work
Item-based collaborative filtering recommendation algorithms
Proceedings of the 10th international conference on World Wide Web
The Eigentrust algorithm for reputation management in P2P networks
WWW '03 Proceedings of the 12th international conference on World Wide Web
Spreading Activation Models for Trust Propagation
EEE '04 Proceedings of the 2004 IEEE International Conference on e-Technology, e-Commerce and e-Service (EEE'04)
SNACK: incorporating social network information in automated collaborative filtering
EC '04 Proceedings of the 5th ACM conference on Electronic commerce
Inferring binary trust relationships in Web-based social networks
ACM Transactions on Internet Technology (TOIT)
Attack-resistant trust metrics for public key certification
SSYM'98 Proceedings of the 7th conference on USENIX Security Symposium - Volume 7
Empirical analysis of predictive algorithms for collaborative filtering
UAI'98 Proceedings of the Fourteenth conference on Uncertainty in artificial intelligence
Alleviating the sparsity problem of collaborative filtering using trust inferences
iTrust'05 Proceedings of the Third international conference on Trust Management
| Hi-index | 0.00 |
Personalization and customization have been shown to be an indispensable function in today's eCommerce businesses and highly applauded by their customers. Collaborative filtering is one of the two major techniques commonly employed by today's recommender systems and has found its way into the recommendation of many diversified types of products. However, collaborative filtering technique suffers from sparsity and cold start problems. The recent emergence of Web 2.0 offers an opportunity to remedy these problems by incorporating the trust relationships explicitly expressed by the users, as evident by some recent research. Previous work in using trust for making recommendation mainly focuses on inferring trust weights for unspecified trust relations. In this paper, we model the problem of using trust for recommendation as a linear program. We then describe two heuristics that leverages trust to estimate the ratings of unseen products by a given user. Finally we develop various strategies of giving continuous trust weights by considering the contextual information pertaining to trust statements and examine their impact on recommendation accuracy using the empirical data collected from Epinion.com. The experimental results show that assigning continuous trust weights using some of the proposed strategies yields higher recommendation accuracy when compared to the baseline approach that gives Boolean trust values.