【作者】 赵军；

【导师】 王强； 王灿；

【作者基本信息】 浙江大学 ， 计算机应用技术， 2011， 硕士

【摘要】 用户越来越习惯在目前流行的论坛社区等网站上进行知识分享,沟通与阅读有趣文章。然而,在论坛大量的内容中,用户却很难在信息过载的情况下找到他们感兴趣的帖子。有两个原因导致传统的个性化推荐系统并不能直接运用在论坛中。一个是论坛网站不像电影与音乐网站那样用户会对产品进行评分；第二个则是由于论坛中用户看帖不回的记录较多,稀疏问题更加严重。另外,检索系统在论坛中的应用也存在问题,在论坛用同一条目检索出的帖子不会使得每个人满意,因此一个个性化的检索也是论坛需要的功能。本文通过挖掘用户回复信息与帖子语义信息,提出了一个可以将用户和帖子映射到同一子空间中表达的算法：用户-帖流形对齐降维。在用户-帖的共有流形中,用户与帖子之间的联系可以方便表达,离目标用户距离最近的就是用户感兴趣的帖子,可以方便地进行推荐。本文还通过用户-帖的流形对齐降维结果对用户兴趣进行建模,进而设计了个性化检索系统,并将用户的反馈信息也集成到检索系统中。在digg. com数据集上做的实验评估表明了本文设计的个性化推荐与检索系统性能非常出色。更多还原

【Abstract】 People are more and more willing to participate in online forums to share their knowledge and experience. However, it may not be easy for them to find their desired threads in online forums due to the information overload problem. Traditional recommendation approaches can not be directly applied to online forums due to two reasons. First, unlike traditional movie or music recommendation problem, there is no rating information in online forums. Second, the sparsity problem is more severe since the users may only read threads but take no-actions. In addition, retrieval system in online forum will probably give results that not interest everybody. Thus, a personalized retrieval answer will be more preferred. To address these limitations, in this paper we propose to make use of the reply relationships among users, as well as thread contents. A learning algorithm is introduced to infer a user-thread alignment manifold in which both users and thread contents can be well represented. Thus, the relatedness between users and threads can be measured on this alignment manifold, and the closest threads which can best meet the corresponding user’s information needs are recommended. With the advance of user-thread alignment manifold, we can also capture user interests as well as user feedbacks to make personalized retrieval system. Experiments on a dataset crawled from digg.com have demonstrated the superiority of our personalized recommendation and retrieval system.更多还原