The anatomy of a large-scale hypertextual Web search engine
WWW7 Proceedings of the seventh international conference on World Wide Web 7
Evaluating collaborative filtering recommender systems
ACM Transactions on Information Systems (TOIS)
The adaptive web
Hi-index | 0.00 |
The online social systems are now playing a more and more important role in our daily life. Information coming from such systems is more personalized and preferable than those from search engines and portals. Those systems are normally described by directed networks where the nodes represent users and the information spreads from leaders to followers. Therefore, the selection of suitable leaders determines the quality of the coming information. In this paper, we propose a leader recommendation method based on a local structure consisting of 4 nodes and 3 directed links. The simulation results on real networks show that our method can accurately recommend the potential leaders. Moreover, further investigation on recommendation diversity indicates that our recommendation method is very personalized. Finally, we remark that our method can be easily extended to improve the existing link prediction algorithms in directed networks.