FRFB: Top-k Followee Recommendation by exploring the Following Behaviors in social networks

摘要

As social networks such asmicro-blogging sites rapidly grow, deciding whom to follow (followeernrecommendation) becomes a significantly important problem. Most existing works exclusivelyrnrely on two traditional factors: the proximity between two users in the network topology orrnthe similarity of the user-generated contents in the social network, disregarding the effect ofrnusers' following behaviors. The challenge ofhowto effectively combine these two factors remainsrnlargely open.Moreover,most research studies simply sort the scores to find top-k users, which isrntime-consuming, especially for large-scale networks. In this paper, we propose the idea that “predictrnusers' following behaviors by following behaviors themselves.”We consider a user's followingrnto others as a normal process of dynamic and coherent behavior, andwemodel the potential propagationrnof the users' following behaviors. Furthermore, based on our previous research on top-krnselection problem, we propose an effective top-k followee recommendation algorithm, calledrnFRFB. FRFB has low complexity and high scalability and, moreover, good adaptability to real-liferndynamic social networks.We conduct extensive experiments, with two real social network datarnsets (Wiki and Twitter), which show that FRFB outperforms the well-known topology-basedrnfollowee recommendation algorithms.