Implicit Social Networks for Social Recommendation of Scholarly Papers

Abstract

Combining social network information with collaborative filtering recommendation algorithms has successfully reduced some of the drawbacks of collaborative filtering and increased the accuracy of recommendations. However, all approaches in the domain of research paper recommendation have used explicit social relations that are initiated by users. Moreover, the results of previous studies have shown that the recommendations produced cannot compete with traditional collaborative filtering. We argue that the available data in social bookmarking websites can be exploited to connect similar users using implicit social connections based on their bookmarking behavior. We explore the implicit social relations between users in social bookmarking websites such as CiteULike and Mendeley, and propose three different implicit social networks to recommend relevant papers/people to users. We showed that the proposed implicit social networks connect users with similar interests and the relations are propagated through the networks. In addition, we showed that implicit social networks connect more users than the two of well-known explicit social networks (co-authorship and friendship).