innovation.gr

This paper studies the problem of recommending new venues to users who participate in location-based social networks (LBSNs). As an increasingly larger number of users par-take in LBSNs, the recommendation problem in this setting has attracted significant attention in research and in practi-cal applications. The detailed information about past user behavior that is traced by the LBSN differentiates the prob-lem significantly from its traditional settings. The spatial nature in the past user behavior and also the information about the user social interaction with other users, provide a richer background to build a more accurate and expressive recommendation model. Although there have been extensive studies on recom-mender systems working with user-item ratings, GPS trajec-tories, and other types of data, there are very few approaches that exploit the unique properties of the LBSN user check-in data. In this paper, we propose algorithms that create rec-ommendations based on four factors: a) past user behavior (visited places), b) the location of each venue, c) the social relationships among the users, and d) the similarity between users. The proposed algorithms outperform traditional rec-ommendation algorithms and other approaches that try to exploit LBSN information. To design our recommendation algorithms we study the properties of two real LBSNs, Brightkite and Gowalla, and analyze the relation between users and visited locations. An experimental evaluation using data from these LBSNs shows that the exploitation of the additional geographical and so-cial information allows our proposed techniques to outper-form the current state of the art