Collaborative Filtering Algorithm Incorporated with Cluster-based Expert Selection ⋆
Weiliang Kong, Qingtang Liu, Zhongkai Yang, Shuyun Han · 2012
In order to solve the scalability and the noise problems suffered by collaborative filtering algorithm, the researchers have proposed expert-based collaborative filtering algorithm. But, there still lacks a principled model for guiding how to select the useful experts. In this paper, firstly, we define a concept of expert which can be reduced into two components: the activity and the influence in a given domain. Secondly, we put forward cluster-based expert selection method. Thirdly, we introduce this method into expert-based collaborative filtering algorithm and propose collaborative filtering algorithm incorporated with cluster-based expert selection. Finally, experiments show that our algorithm has better performance than the existing expert-based collaborative filtering algorithm on recommendation precision (about 12% improvement) and predication accuracy (about 1.8% improvement).