• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

J4 ›› 2015, Vol. 37 ›› Issue (01): 179-183.

• 论文 • 上一篇    下一篇

基于用户人口统计与专家信任的协同过滤算法

焦东俊   

  1. (北京邮电大学智能通信软件与多媒体北京市重点实验室,北京 100876)
  • 收稿日期:2014-09-13 修回日期:2014-11-16 出版日期:2015-01-25 发布日期:2015-01-25

Collaborative filtering algorithm based on
user demographics and expert opinions 

JIAO Dongjun   

  1. (Beijing Key Laboratory of Intelligent Telecommunications,
    Beijing University of Posts and Telecommunications,Beijing 100876,China)
  • Received:2014-09-13 Revised:2014-11-16 Online:2015-01-25 Published:2015-01-25

摘要:

推荐系统是学术研究的热门课题,在工业界应用也越来越广泛,推荐系统旨在为用户推荐相关的感兴趣的物品。协同过滤算法被用来比较用户及物品的相似度,向用户推荐与其最近邻用户的偏好。为了提高协同过滤算法预测的准确率,提出基于用户人口统计与专家信任的协同过滤算法,先比较用户人口统计属性,然后进一步比较用户与专家的人口统计属性来得到用户与专家的相似度,从而提高预测的准确性。实验验证表明,提出的算法能够有效提高协同过滤算法预测的准确率。

关键词: 推荐系统, 协同过滤算法, 人口统计, 专家信任

Abstract:

Recommender systems have been used tremendously in academia and industry, and the recommendations generated by these systems aim to offer relevant interesting items to users. Collaborative filtering algorithm is used to calculate the similarities between users and items, and recommends the nearest neighbors’ preferences to users. In order to improve the prediction accuracy of collaborative filtering algorithm, we propose a collaborative filtering algorithm based on user demographics and expert opinions. First we compare users’ demographic attributes, which are then compared with expert demographic attributes to calculate the similarities between users and experts. Experimental results verify that the algorithm proposed in this paper can effectively improve the prediction accuracy of collaborative filtering algorithm.

Key words: recommender system;collaborative filtering;demographic correlation;expert opinions