Computer Engineering and Applications ›› 2011, Vol. 47 ›› Issue (13): 154-156.

• 数据库、信号与信息处理 • Previous Articles     Next Articles

Exponential forgetting collaborative filtering recommendation algorithm incorporated with user interest change

LI Kechao,LIANG Zhengyou   

  1. School of Computer,Electronics and Information,Guangxi University,Nanning 530004,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2011-05-01 Published:2011-05-01

适应用户兴趣变化的指数遗忘协同过滤算法

李克潮,梁正友   

  1. 广西大学 计算机与电子信息学院,南宁 530004

Abstract: Collaborative Filtering(CF) is the most successful technologies to date,but traditional collaborative filtering algorithm does not consider the change of users’ interests.To solve the problem,this paper presents a CF recommendation algorithm,named item similarity-based CF algorithm,which uses exponential gradual forgetting-based data weight to diminish the importance of each item similarity.The experimental results show that the proposed algorithm can efficiently improve recommendation quality.

Key words: collaborative filtering, recommendation algorithm, users’ interests, exponential forgetting

摘要: 协同过滤是目前推荐系统中最为成功的推荐技术,但传统的协同过滤算法没有考虑用户兴趣的变化。针对上述问题,从艾宾浩斯记忆遗忘规律得到启发,提出一种基于资源相似度的协同过滤算法,利用基于指数遗忘的数据权重来逐步减小资源相似度的权重。实验结果表明,该算法显著提高推荐系统的推荐质量。

关键词: 协同过滤, 推荐算法, 用户兴趣, 指数遗忘