A Collaborative Filtering Recommendation Algorithm Based on Density Peak Clustering
Zhihe Wang, Teng Zhang, Hui Du · 2019
With the increasing of data, the recommendation performance of the traditional collaborative filtering recommendation algorithm is getting lower. In this paper, a new clustering algorithm is proposed, which is combined with collaborative filtering recommendation to improve the recommendation efficiency. The proposed algorithm uses density peak clustering (FDP) to preliminarily determine the clustering center as the initial center of K-means algorithm which solves the disadvantage that K-means algorithm is easy to fall into local optimization. The clustering algorithm is used to divide users, and then collaborative filtering recommendation is carried out according to the category to which the target user belongs, which provides users with more "personalized" recommendation. The experimental results show that compared with K-means, the proposed clustering algorithm improves the clustering accuracy. In addition, the clustering algorithm combined with collaborative filtering recommendation is verified on Group Lens-Movie Lens dataset, and the results show that the algorithm is more effective than the traditional single cooperative filtering recommendation algorithm.