An Improved Singular Value Decomposition Recommender Algorithm Based on User Trust Relationship
Pan Junchi, Xingming Zhang, Xiaofeng Qi · 2015
Collaborative filtering is one of the most widely used recommender algorithms, whereas it is suffering the issues of data sparsity.Recommender algorithms based on trust perform better in alleviating data sparsity.However, there remain shortages in the process of mining trust relation in specific algorithms, which limit the improvement of prediction accuracy.To address this problem, the paper proposes an improved singular value decomposition algorithm, trying to integrate truster-specific and trustee-specific information and the implicit feedback of each when generating predictions.Experiments on the Epinions dataset show that the proposed algorithm performs better than state-of-the-art recommender algorithms in prediction accuracy.