Collaborative filtering with social regularization for TV program recommendation

In recent years, we have witnessed the explosive growth of microblogging services. As a popular platform for users to communicate and share information with friends, microblog has opened up new opportunities for recommendation. In this paper, we explore the possibility of recommending TV programs with microblogs. In particular, we leverage the following two important features of microblogs: (1) the rich user generated content reveals users' preferences on TV programs as well as the properties of TV programs and (2) the social interactions of the users suggest the mutual influences among the users. Taking into consideration of the above two properties, we proposed a hybrid recommendation model based on probabilistic matrix factorization, a popular collaborative filtering method. Two regularizers are added during matrix factorization: the social regularizer and the item similarity regularizer. We validate the proposed algorithm with Sina Weibo data set for TV program recommendation. The experimental results show that the proposed algorithm significantly outperforms the state-of-the-art collaborative filtering method, demonstrating the importance of incorporating social trust and item similarity in recommendation. In addition, we show that the proposed method is robust in recommending to new users, a typical cold-start scenario.