Missing Value Prediction Using Co-clustering and RBF for Collaborative Filtering

Youchun Ji, Wenxing Hong, Jianwei Qi

Abstract: The sparsity of user-item rating matrix will reduce the performance of collaborative filtering algorithm in news recommendation system. In order to overcome the problem, we predict the values of user-item rating matrix combining two approaches: co-clustering and Radial Basis Function network (RBF). Co-clustering algorithm simultaneous cluster the rows and columns of the user-item rating matrix. It can cluster the matrix into several small matrix with high similarity. We take advantage of the similarity of a cluster and then predict the values using RBF network. This method can complement the sparse user-item rating matrix and improve the accuracy of news recommendation system by collaborative filtering algorithm. Experimental results on Xiamen University campus news data set demonstrate the efficiency and effectiveness of the proposed missing value prediction method.