Abstract

  Recommender systems attempt to suggest information that is of potential interest to users helping them to quickly find information relevant to them. In addition to historical user–item interaction data, such as users’ ratings on items, social recommendation methods use social relationships between users to improve the accuracy of recommendations. However, the available social relationships are often extremely sparse. Therefore, incorporating implicit relationships into the recommendation process can be effective to improve the performance of social recommender systems, especially for those users whose explicit relationships are insufficient to make accurate recommendations. The existing approaches have not considered reliability of the implicit relationships. In this paper, a social recommender system is proposed based on reliable implicit relationships. To this end, Dempster–Shafer theory is used as a powerful mathematical tool to calculate the implicit relationships. Moreover, a new measure is introduced to evaluate the reliability of predictions, where unreliable predictions are recalculated using a neighborhood improvement mechanism. This mechanism uses a confidence measure between the users to identify ineffective users in the neighborhood set of a target user. Finally, new reliable ratings are calculated by removing the identified ineffective neighbors. Extensive experiments are conducted on three well-known datasets, and the results demonstrate that our approach achieves superior performance to the state-of-the-art recommendation methods.