Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
5102485 | Physica A: Statistical Mechanics and its Applications | 2018 | 12 Pages |
Abstract
Recommendation algorithms based on bipartite networks have become increasingly popular, thanks to their accuracy and flexibility. Currently, many of these methods ignore users' negative ratings. In this work, we propose a method to exploit negative ratings for the network-based inference algorithm. We find that negative ratings play a positive role regardless of sparsity of data sets. Furthermore, we improve the efficiency of our method and compare it with the state-of-the-art algorithms. Experimental results show that the present method outperforms the existing algorithms.
Keywords
Related Topics
Physical Sciences and Engineering
Mathematics
Mathematical Physics
Authors
Liang Hu, Liang Ren, Wenbin Lin,