Geographical and temporal similarity measurement in location-based social networks
Zhengwu Yuan, Yanli Jiang, Győző Gidófalvi · 2013
Using "check-in" data gathered from location-based social networks, this paper proposes to measure the similarity of users by considering the geographical and the temporal aspect of their geographical and temporal aspects of their "check-ins". Temporal neighborhood is added to support the time dimension on the basis of the traditional DBSCAN clustering algorithm, which determines the similarity among users at different scales using the classical Vector Space Model (VSM) with vectors composed of the amount of visits in different cluster area. The spatio-temporal similarity of the user behaviors are obtained through overlapping the different weighted user similarity values. The experimental results show that the proposed approach is effective in measuring user similarity in location-based social networks.