tailieunhanh - Báo cáo khoa học: "User Participation Prediction in Online Forums"

Online community is an important source for latest news and information. Accurate prediction of a user’s interest can help provide better user experience. In this paper, we develop a recommendation system for online forums. There are a lot of differences between online forums and formal media. For example, content generated by users in online forums contains more noise compared to formal documents. Content topics in the same forum are more focused than sources like news websites. | User Participation Prediction in Online Forums Zhonghua Qu and Yang Liu The University of Texas at Dallas qzh yangl@ Abstract Online community is an important source for latest news and information. Accurate prediction of a user s interest can help provide better user experience. In this paper we develop a recommendation system for online forums. There are a lot of differences between online forums and formal media. For example content generated by users in online forums contains more noise compared to formal documents. Content topics in the same forum are more focused than sources like news websites. Some of these differences present challenges to traditional word-based user profiling and recommendation systems but some also provide opportunities for better recommendation performance. In our recommendation system we propose to a use latent topics to interpolate with content-based recommendation b model latent user groups to utilize information from other users. We have collected three types of forum data sets. Our experimental results demonstrate that our proposed hybrid approach works well in all three types of forums. 1 Introduction Internet is an important source of information. It has become a habit of many people to go to the internet for latest news and updates. However not all articles are equally interesting for different users. In order to intelligently predict interesting articles for individual users personalized news recommendation systems have been developed. There are in general two types of approaches upon which rec ommendation systems are built. Content based recommendation systems use the textual information of news articles and user generated content to rank items. Collaborative filtering on the other hand uses co-occurrence information from a collection of users for recommendation. During the past few years online community has become a large part of internet. More often latest information and knowledge appear at online community