Locations recommendation based on check-in data from Location-Based Social Network

In recent years, together with the universal use of GPS embedded mobile phones and popularity of social network, Location-Based Social Network (LBSN) has been a hit, and user volume rises continuously. The prevalent of LBSN contributes massive data for pattern recognition and behavior analysis. In this paper we mainly discuss location recommendation based on check-in data of LBS. Four feasible methods have been proposed, with respect to both content-based and collaborative filtering algorithms. The methods we put forward base on models such as standard deviation ellipse, buffer, topology as well as utility matrix and all these models perform well and satisfying in location recommendation.