An Improved Model of Trust-aware Recommender Systems Using Distrust Metric

Abstract

Trust-aware recommender systems are intelligent technology applications that make use of trust information and user personal data in social networks to provide personalized recommendations. Recent research on recommender systems shows that these recommender systems are more robust against shilling attacks and can better be used for generating recommendations for new users. In this paper we proposed a model for improving the accuracy of trust-aware recommender systems. The results of evaluating our approach on Extended Epinions dataset shows that this approach can improve accuracy of recommender systems significantly while does not reduce the coverage of recommender systems.