一种改进的协同过滤推荐算法  

An Adaptive Algorithm for Collaborative Filtering Recommendation

作　　者：杨芳[1] 潘一飞[2] 李杰[1] 王云峰[1] 

机构地区：[1]河北工业大学管理学院,天津300401 [2]四川大学电气信息学院,四川成都610065

出　　处：《河北工业大学学报》2010年第3期82-87,共6页Journal of Hebei University of Technology

基　　金：河北省自然科学基金(F2008000117);河北省科技攻关项目(07213508D)

摘　　要：电子商务的蓬勃发展,使网站中能够提供的商品种类日益繁多,如何迎合客户的兴趣来推荐商品,成为当前电子商务亟待解决的重点问题.协同过滤作为目前推荐系统应用中最为成功的个性化推荐技术,也得到了越来越多研究者的关注.文章在简要介绍传统协同过滤推荐算法的基础上,重点对推荐算法无法适用于用户多兴趣下的推荐问题进行了剖析,提出了一种基于用户多兴趣的协同过滤推荐改进算法.通过实验仿真,验证了该算法的有效性.The vigorous development of e-commerce has enabled websites to provide an increasingly larger variety of products. How to recommend products catering for the interest of customers has become an important issue that urgently requires today＇s e-commerce to solve. Collaborative filtering, as the most successful personalized technique of recommendation in the existing recommendation system application, has gained more and more researchers＇ attention. Based on a brief introduction of traditional collaborative filtering recommendation, the paper is focused on analysis ofthe problem that recommendation algorithm fails to apply to recommendation with the users＇ multiple interests, and proposes an improved algorithm of collaborative filtering based on multiple interests of users. Via experimental simulation, the validity of the algorithm is verified.

关 键 词：电子商务 个性化推荐 数据挖掘 协同过滤推荐算法 用户多兴趣 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论]