Абстрактный

Combining social network information with probabilistic matrix factorization to enhance recommendation performance

Hui Li , Yun Hu , Jun Shi, Yong Zhang

This paper examines the problem of social collaborative filtering to recommend items of interest to users in a social network setting. Many social networks capture the relationships among the nodes by using trust scores to label the edges. The bias of a node denotes its propensity to trust/mistrust its neighbours and is closely related to truthfulness. It is based on the idea that the recommendation of a highly biased node should be removed. In this paper, we propose a model-based approach for recommendation employing matrix factorization after removing the bias nodes from each link, which naturally fuses the users’ tastes and their trusted friends’ favours together. The empirical analysis on real large datasets demonstrate that our approaches outperform other state-ofthe- art methods