Research of hybrid collaborative filtering algorithm based on news recommendation

Abstract

This paper introduced the personalized recommendation technology to the news system. In order to meet the demand of the users' personality and ease the problem of data sparse, the research work proposed the hybrid collaborative filtering algorithm based on news recommendation. It improved correlation coefficient formula by adding news hot parameter when calculating the similarity of users, and then used hybrid recommendation algorithm to forecast users' ratings to make user-rating matrix to non-zero values. Experimental results illustrated that the hybrid recommendation algorithm can effectively increase the accuracy and stability of recommendation so as to achieve better recommendation results.