With rapid development of wireless communication technologies, such as global position system, location-based social networks (LBSNs), like Foursquare, Facebook, etc., have attracted millions of users to share their social friendship and locations via check-in. As one of the most important tasks in LBSNs, POI recommendation aims to mining user’s preference on locations and to provide recommendations to users based on the plenty of check-in information. In this work, we propose to use tensor factorization to handle this problem, a three-mode tensor is used to model all user’s check-in behavior, then CP decomposition is applied to tensor factorization and to recovery the original tensor. We conduct some experiment on a large-scale real-word LBSNs. I will also show our future work on POI recommendation.