Application of game theory techniques for improving trust based recommender systems in social networks

Recommender system is a solution to the information overload problem in websites that allow users to express their interests about items. Collaborative filtering is one of the most important methods in recommender systems which predicts ratings for active user based on opinions and interests of other users who are similar to the active user. Accuracy of ratings prediction can be considerably improved using trust statements between users in recommender systems. In this paper, a novel method is proposed to determine effectiveness coefficient of the users in trust network of the active user. For this purpose, the Pareto dominance concept is used to identify dominance users of the active user and the trust statements between users are calculated based on this concept. Experimental results on Epinions dataset show that the proposed method improve accuracy of ratings prediction while provide suitable coverage rather than several well-known state-of-the-art methods.