Bookmark recommendation in social bookmarking services using Wikipedia

Abstract

Social bookmarking systems allow users to attach freely chosen keywords as tags to bookmarks of web pages. These tags are used to recommend relevant bookmarks to other users. However, there is no guarantee that every user get enough bookmark recommended, because of the diversity of tags. In this paper, we propose a personalized recommender system using Wikipedia. Our system extends a tag set to find similar users and relevant bookmarks by using the Wikipedia category database. The experimental results show that significant increase of relevant bookmarks recommended without notable increase of the noise.