推荐系统中的用户兴趣模式检测
首发时间:2008-12-18
摘要:近年来推荐系统已经被证明是一种非常有效的方法用以处理网络上的信息过载问题,这些推荐系统根据用户以往的访问记录为用户兴趣建模,并根据用户兴趣给用户提供推荐项。然而,当前在推荐系统方面的研究几乎都忽略了用户访问记录的时间因素,这些系统在为用户做推荐的时候并不知道用户的兴趣是否发生了改变,或者说并不知道在推荐那一刻并不知道用户真正需要什么。本文对用户访问记录进行深入分析,提出了四种典型的用户兴趣模式,并提出了一种及于子图稠密度和连续度的方法来检测这四种模式。实验结果显示该方法可以有效检测用户兴趣模式。
For information in English, please click here
Detecting Interest Pattern of User In Recommender System
Abstract:Recommender systems have been a success in the past several years as an effective way to help people cope with the problem of information overload. These systems produce recommendations for the users according to their interests represented by their past behavior. However, the current research on recommender systems has paid little attention to the use of time related data in the recommendation process. And these systems don’t know if the interest of one user has changed or what does one user really need at that recommending time. This paper suggests four typical interest patterns, and proposes a methodology for detecting a user’s time-variant interest pattern with a heuristic segmentation method applied. Experimental results show that the approach proposed is effective to detecting user interest patterns.
Keywords: recommender system interest pattern shifting detecting series segmentation
论文图表:
引用
No.2673327730812295****
同行评议
勘误表
推荐系统中的用户兴趣模式检测
评论
全部评论0/1000