What should I comment: Recommending posts for commenting

Nowadays, with the appearance of the Internet and personal computers, Web becomes one of the most important vehicles to convey information. There are many new forms of information on the Web, including websites, blogs, wikis, social networks, and Internet forums. The explosion of user-generated content poses challenges to browsing and finding valuable information on the Web. In this paper, we present a study on the task of recommending, for a given user, a short list of suitable forum posts for commenting. We propose a collaborative filtering method which exploits the co-commenting patterns of the users to generate recommendations, and compare the method with traditional content-based filtering approaches. Experimental results on two types of forums show that the proposed collaborative filtering method achieved substantial improvements in terms of accuracy over a baseline and the content-based filtering methods. The results also demonstrate the stability of our method in handling new posts with small number of comments.