华东师范大学学报(自然科学版) ›› 2016, Vol. 2016 ›› Issue (3): 60-66.doi: 10.3969/j.issn.1000-5641.2016.03.007

• 计算机科学 • 上一篇    下一篇

基于用户相似性的协同过滤算法改进

 王威, 郑骏   

  1. 华东师范大学 计算中心,上海200062
  • 收稿日期:2015-05-22 出版日期:2016-05-25 发布日期:2016-09-22
  • 通讯作者: 郑骏,男,教授,主要研究方向为Web开发及应用. E-mail: jzheng@cc.ecnu.edu.cn.
  • 作者简介:王威,男,硕士研究生,主要研究方向为Web开发及应用.
  • 基金资助:

    国家高技术研究发展计划(2013AA01A211)

Improved collaborative filtering algorithm based on usersimilarity

 WANG  Wei, ZHENG  Jun   

  • Received:2015-05-22 Online:2016-05-25 Published:2016-09-22

摘要: 协同过滤技术作为目前最常见的个性化推荐技术之一,被广泛认可和应用.作为基于内容的算法执行方式,协同过滤在准确性上具有相当的优势.该算法的核心问题是相似度的计算.本论文介绍了传统协同过滤算法,并对原有的相似度公式进行了优化,使得相似度计算更具有准确性.实验表明,文中提出的优化方法在推荐精度上有显著提高,降低了平均绝对误差(Mean Absolute Error, MAE).

关键词: 推荐技术, 相似度, 协同过滤, MAE

Abstract: Collaborative filtering is widely accepted and applied currently as one of the most popular personalized recommendation methods. It is an implementation method based on content that has considerable advantages in accuracy. The core issue of collaborative filtering is how to work out the calculation of similarity. In this paper, we introduce the traditional collaborative filtering algorithm and make similarity calculation more accurately by optimizing the traditional formula of similarity. Experimental results show that the optimized algorithm can improve the accuracy of the recommendation and reduce the MAE (Mean Absolute Error, MAE) efficiently.

Key words: recommendation methods, similarity, collaborative filtering, MAE

中图分类号: