Hits:

Indexed by:期刊论文

Date of Publication:2013-01-01

Journal:Information (Japan)

Included Journals:Scopus

Volume:16

Issue:7 A

Page Number:4555-4576

ISSN No.:13434500

Abstract:Collaborative filtering is one of the most successful recommending techniques, but it suffers from the problem of "information overload". Meanwhile, it is assumed in traditional collaborative filtering methods that all users have the same weight on ratings, thus they are unable to distinguish users who have similar tastes but different rating behaviors. Besides, collaborative filtering doesn't consider that customers' interests and demands may vary with contexts in mobile environment These problems severely affected the quality of recommendation. To solve those problems, we propose a novel collaborative filtering method combining context clustering and social network analysis. Firstly, all users are clustered into different groups by context information to reduce the sparsity and dimension of ratings data. Then, a user ranking model based on social network analysis is constructed to estimate the correlations between users, and incorporated into similarity measure for improving the quality of recommendation. Experiments on three real-world datasets are carried out to evaluate the performance of our method. The results show that the proposed method outperforms other methods and improves recommendation quality effectively. ? 2013 International Information Institute.