Using multi-features to recommend friends on location-based social networks.

Location-based social networks (LBSNs) not only offer novel services but also produce more abundant data to help new serves for human. It will help discover latent trajectory, possible friendship and then guide trip, predict next place, recommend friends, and promote sales and so on. In this paper we study two problems for friend recommendation on LBSN: what is the main feature and how to predict friendship? We firstly analyze many factors related with human mobility and social relations; adopt the information gain to measure the contribution of different features to human friendship. Then we extract user social relationship, check-in distance during fixed periods and check-in type as key features. Because the prediction problem could be considered as a classification problem, we choose SVM to predict friendship. At last some experiment results show our algorithm valid to some extent. [ABSTRACT FROM AUTHOR]

Copyright of Peer-to-Peer Networking & Applications is the property of Springer Nature and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)