Analysis of Trust-Based E-Commerce Recommender Systems Under Recommendation Attacks
Fuguo Zhang, Xu Sheng-hua · 2007
Recommender systems have been accepted as a vital application on the Web by offering product advice or information users might be interested in. However, conventional collaborative filtering recommender systems are highly vulnerable to attacks. Malicious users can inject a large number of biased profiles into such a system in order to make recommendations that favor or disfavor given items. In this paper, we assess the robustness of our topic-level trust-based recommendation algorithm that incorporate topic-level trust model into classic collaborative filtering algorithm. We perform a series of experiments to quantitatively evaluate the effect of two popular recommendation attacks on the topic-level trust based algorithm by comparing with traditional user-based collaborative filtering algorithm and profile-level trust based recommender algorithm. The results show that topic-level trust based collaborative filtering algorithm offers significant improvements in stability and robustness over the standard k-nearest neighbor approach when attacked.