A New Collaborative Filtering Algorithm with Combination of Explicit Trust and Implicit Trust
Huan Zhao, Yuansun Zhang, Yufeng Xiao · 2018
To alleviate data sparsity and cold start problems of conventional collaborative filtering, social trust information has been incorporated into recommender systems. There exist a few approaches, which mainly use the trust scores explicitly expressed by users and effectively improve accuracy of rating prediction. However, having users provide explicit trust scores of each other is often full of challenges. In this paper, we propose a novel method to incorporate both explicit and implicit trust in providing recommendations. First, we use one of Trust Metrics algorithms to compute and predict implicit trust scores between users based on their interactions. Then, we merge ratings of a user's explicitly and implicitly trusted neighbors to complement and represent the preferences of the user. Experimental results show that our method outperforms other counterparts both in terms of accuracy and coverage.