The rapid growth of location-based services(LBSs)has greatly enrichedpeople's urban lives and attracted millions of users in recent years.Location-based social networks(LBSNs)allow users to check-in at a physicallocation and share daily tips on points-of-interest (POIs) with their friendsanytime and anywhere. Such check-in behavior can make daily real-lifeexperiences spread quickly through the Internet. Moreover, such check-in datain LBSNs can be fully exploited to understand the basic laws of human dailymovement and mobility. This paper focuses on reviewing the taxonomy of usermodeling for POI recommendations through the data analysis of LBSNs. First, webriefly introduce the structure and data characteristics of LBSNs,then wepresent a formalization of user modeling for POI recommendations in LBSNs.Depending on which type of LBSNs data was fully utilized in user modelingapproaches for POI recommendations, we divide user modeling algorithms intofour categories: pure check-in data-based user modeling, geographicalinformation-based user modeling, spatio-temporal information-based usermodeling, and geo-social information-based user modeling. Finally,summarizingthe existing works, we point out the future challenges and new directions infive possible aspects
展开▼