Information filtering via self-consistent refinement

被引：70

作者：

Ren, Jie ^{[1
,2
,3
]}

Zhou, Tao ^{[1
,4
,5
]}

Zhang, Yi-Cheng ^{[1
,5
]}

机构：

[1] Univ Fribourg, Dept Phys, CH-1700 Fribourg, Switzerland

[2] Natl Univ Singapore, Dept Phys, Singapore 117542, Singapore

[3] Natl Univ Singapore, Ctr Computat Sci & Engn, Singapore 117542, Singapore

[4] Univ Sci & Technol China, Dept Modern Phys, Hefei 230026, Peoples R China

[5] Univ Elect Sci & Technol China, Informat Econ & Internet Res Lab, Chengdu 610054, Peoples R China

来源：

EPL | 2008年 / 82卷 / 05期

关键词：

D O I：

10.1209/0295-5075/82/58007

中图分类号：

O4 [物理学];

学科分类号：

0702 ;

摘要：

Recommender systems are significant to help people deal with the world of information explosion and overload. In this letter, we develop a general framework named self-consistent refinement and implement it by embedding two representative recommendation algorithms: similarity-based and spectrum-based methods. Numerical simulations on a benchmark data set demonstrate that the present method converges fast and can provide quite better performance than the standard methods. Copyright (c) EPLA, 2008.

引用

页数：5

共 32 条

[1] Toward the next generation of recommender systems: A survey of the state-of-the-art and possible extensions [J].