基于信任和兴趣聚类的协同过滤算法

With the development of Web 2.0 and kinds of social networking sites, personalized services in social network received more and more attention. In order to provide high quality social recommendation, a collaborative filtering algorithm based on trust and interest clustering is proposed in this paper. Firstly, the algorithm takes synthetieally into account users' trust degree and the similarity of interest and users are clustered based on the composite similarity. Then we find out the nearest neighbours of target user in his cluster and predict the unknown user rates. Finally, experimental results show that the proposed method can obtain more accurate social recommendation as well as improving the real-time performance to some extent.