A collaborative recommendation algorithm based on user cluster classification
Zihao Chen, Zhengmin Li · 2016
Recommendation system is playing an increasingly important role for Internet service, the information overload is one of the most crucial event that users encounter on the Internet. Personalized recommendation system is an effective technique to solve the current problem. However, with the development of collaborative recommendation with heterogeneous explicit feedbacks such as 5-star grade scores rating systems, the difficulties of extreme sparsity of user rating data have become more and more severe. In this paper, we proposed a collaborative recommendation algorithm based on user cluster classification (UCC). In order to obtain a more realistic users similarity, we propose a method to increase the accuracy of the score prediction by improve the similarity between the effective user. Experimental results show that, the proposed algorithm achieved an average of 5.73% improvement in terms of recommended quality using sparse data and the small neighboring environment.