通信学报 ›› 2016, Vol. 37 ›› Issue (1): 199-207.doi: 10.11959/j.issn.1000-436x.2016024

• 学术通信 • 上一篇    

基于用户模糊相似度的协同过滤算法

吴毅涛,张兴明,王兴茂,李晗   

  1. 国家数字交换系统工程技术研究中心,河南 郑州 450002
  • 出版日期:2016-01-25 发布日期:2016-01-27
  • 基金资助:
    国家重点基础研究发展计划(“973”计划)基金资助项目;国家高技术研究发展计划(“863”计划)基金资助项目

User fuzzy similarity-based collaborative filtering recommendation algorithm

tao WUYi,ming ZHANGXing,mao WANGXing,Han LI   

  1. National Digital Switching System Engineering and Technological R&D Center, Zhengzhou 450002, China
  • Online:2016-01-25 Published:2016-01-27
  • Supported by:
    The National Basic Research Program of China(973 Program);The National High Tech-nology Research and Development Program of China (863 Program)

摘要:

针对离散评分不能合理表达用户观点和传统协同过滤算法存在稀疏性等问题,借鉴年龄模糊模型,提出了梯形模糊评分模型。该模型将离散评分模糊化为梯形模糊数,考虑了评分模糊性和信息量,通过梯形模糊数来计算用户相似度,据此设计了协同过滤算法,并证明了该算法是传统协同过滤算法在模糊域的扩展。实验表明,该算法在数据稀疏且用户数远多于项目数时性能突出,并且算法运行时间远小于传统协同过滤算法。

关键词: 协同过滤, 梯形模糊评分模型, 模糊距离, 模糊相似度

Abstract:

In order to reflect the actual case of human decisions and solve the data sparseness problem of traditional col-laborative filtering recommendation algorithm, a trapezoid fuzzy model based on age fuzzy model was proposed. In this model, crisp point was fuzzified into trapezoid fuzzy mber and the fuzziness and information of users' grade was taken into account when calculating user's similarity by trapezoid fuzzy number. Based on this model, the user fuzzy similari-ty-based collaborative filtering recommendation algorithm was designed. The algorithm was proved to be an extension of traditional collaborative filtering algorithm in fuzzy fields. The experimental results show that, the proposed algorithm performs better when implemented in the sparse dataset with more user than item, and its running time is much less than traditional collaborative filtering algorithm.

Key words: collaborative filtering, trapezoid fuzzy model, fuzzy distance, fuzzy similarity

No Suggested Reading articles found!