A New Users Interests Prediction Method Based on Social Hub in Microblog

Abstract

As a significant mean for users sharing useful information, microblogs have played an important role in people daily life. People are overwhelmed in a large amount of daily updated microblog information. It becomes increasing necessary for social network users to be recommended useful information of interest to them. In order to meet the demand of users we put forward a new method for user interest prediction. Previous works have focused on analyzing user’s social relations and time information by observing users’ history behavior. We extend currently approach by incorporating social hub information, such as social relations, topic correlations, and social hub interest trend and so on. Our performance evaluation by Pren demonstrates the effectiveness of the proposed model and the experimental results show that it can accurately predict uses’ interests.