摘要
个性化推荐技术在电子商务系统中得到了广泛应用.针对现有的用户模型不能及时根据用户自身兴趣偏移进行更新的问题,提出了一种基于用户行为反馈的兴趣度模型的更新算法,在创建好模型的基础上,分析用户的购买记录和用户的浏览行为,结合用户的兴趣内容,实现用户兴趣的自动更新,得到的针对新的用户兴趣的推荐商品列表,在此基础上结合用户的购买商品记录,实现推荐商品的个性化排序,从而向用户进行个性化推荐.实验对比结果表明,该算法能更好地发现用户当前的购买兴趣,从而进一步提高个性化推荐精度和用户满意度.
Personalized recommendation is a widely applied technology in e-commerce.Since the existing user models can not renew in time from the consumer interest changing,the paper presents a personal interest degree model Updating algorithm based on consumer behavior feedback,which analyzes consumer purchase history and consumer behavior mode,and updates the consumer interest automatically from the user browsed contents to get the recommended list.On this basis,the algorithm can predict the personal intererst order of recommended commodities from the purchase history,which can be used to make the personalized recommendation for each user.Experimental results show that the algorithm can identify user personalization interest more efficiently,since it can improve the recommendation accuracy and customer satisfaction.
出处
《辽宁大学学报（自然科学版）》
CAS
2011年第1期40-45,共6页
Journal of Liaoning University：Natural Sciences Edition
关键词
电子商务
个性化推荐
兴趣度模型
用户行为反馈
E-commerce
personalized recommendation
interest degree model
consumer behavior feedback
