Producing Friend Recommendations in a Social Bookmarking System by Mining Users Content

Social Bookmarking Systems (and Social Media Systems in general) are experiencing a quick growth in the number of active users. This expansion led to the well-known “social interaction overload” problem, that means that each user has too many potential people to interact with. In order to address this problem, user recommender systems are widely proposed in the social media literature to recommend friends or people to follow. Currently, there are no approaches able to produce friend recommendations in the Social Bookmarking Systems domain. In this paper we propose a friend recommendation algorithm for a Social Bookmarking System, based on low computational effort heuristics that allow real time applications. Experimental results show that, when users tag in the same way and are also interested in the same content, they can be recommended as friends. The proposed algorithm produces better results, with respect to policies that use only tags and do not consider content.