Heterogeneous trust-aware recommender systems in social network

Trust, as the basis of human interactions, has been playing an important role in addressing information sharing, experience communication, and public opinions. Trust-aware recommender systems are an effective solution to the information overload problem, especially in the online world where we are constantly faced with inordinately many choices. In this paper, to build a trust-aware recommender system with enhanced accuracy of recommendation, a novel approach is proposed which incorporates multi-faceted trust relationships between users into traditional rating prediction algorithms to reliably estimate users multi-faceted and asymmetry trust strengths. Experimental results on real-world data show that our work of discerning heterogeneous trust can be applied to improve the performance rating prediction and more robust to the cold-start problem.