Research On The Method Of Point Of Interest Recommendation In Location-Based Social Network

In recent years,with the rapid development of mobile Internet technology,positioning technology,wireless sensor technology and the popularity of smart phones,location-based social networks(LBSNs)and its application services have developed rapidly.Location data bridges the gap between the physical and digital worlds and enables a deeper understanding of users' preferences and behaviors.Since users generate a large amount of check-in data in LBSNs,it is possible to recommend points of interest(POIs)to users.POI recommendation can help users understand unfamiliar cities and help them choose their destination.And POI recommendation is also of great commercial value.This makes POI recommendation become an important service of LBSNs,which has widely received attention in both academia and industry.Currently,POI recommendation is becoming a new research hotspot in the fields of recommendation system and social networks.Compared to traditional recommendation,POI recommendation becomes more complex due to the dependence of users and locations and the unique attributes of locations in LBSNs.POI recommendation faces many challenges that differ from traditional recommendation,such as implicit user feedback,data sparsity,contextual diversity and so on.In order to cope with these challenges,this study proposes a general POI recommendation method based on spatial and mobility pattern using collaborative filtering.General POI recommendation only recommends POIs that a user may access in the future.And it cannot recommend POIs that a user may access at the next moment according to the user's current context information,such as the user's current time and current location.Therefore,general POI recommendation cannot make personalized recommendation for a user according to the user's current context information.In order to solve this problem,this study proposes a next POI recommendation method that integrates geospatial and temporal preference.Since the above two types of POI recommendations do not take into account whether a user has previously visited the POIs recommended to the user,this can affect the novelty of the recommendation and the user's experience.In order to provide users with a more practical and experienceable POI recommendation,this study proposes a next new POI recommendation based on simulated user travel decision-making process.The main work and contributions of this paper are summarized as follows:(1)For the general POI recommendation,this study proposes a general POI recommendation method based on spatial and mobility pattern using collaborative filtering,called SMPCF.First,SMPCF mines the target user's active area based on his or her check-in history,and designs a personalized user spatial similarity calculation method based on the target user's active area.Secondly,SMPCF takes into account three features of the human mobility pattern: spatial,temporal,and sequential properties.Furthermore,SMPCF design a personalized user mobility pattern similarity calculation method based on the features of human mobility pattern.Finally,a recommendation list is generated based on the idea of collaborative filtering.Compared with the state-ofthe-art POI recommendation approaches,the experimental results demonstrate that SMPCF achieves much better performance.(2)For the next POI recommendation,this study proposes a next POI recommendation method that integrates geospatial and temporal preferences,called IGTP.Compared with general POI recommendation,IGTP can provide more personalized recommendation for users according to their context information.IGTP has the following advantages: First,IGTP uses the classification information of POI to model users' check-in history to effectively overcome the sparsity of the data.Secondly,IGTP takes into account the geographical distance and density factors that affect people's choice of POI,and limits POIs to be recommended to the potential activitive area centered on the current location of the target user.Finally,IGTP integrates geospatial and user preferences into a unified recommendation framework.Compared with the state-of-the-art next POI recommendation approaches,the experimental results demonstrate that IGTP achieves much better performance.(3)For the next new POI Recommendation,this study proposes a next new POI recommendation method,called STDMP.STDMP simulates a user's travel decisionmaking process by weighing two important factors that affect a user's travel decision: preference factors and geographic factors.First,STDMP use tensor to model user's check-in history and dynamically predict user preferences.Then,in order to characterize the influence of geographic factor on individual users,STDMP designs a personalized user similarity calculation method and fits curves for the target user to reflect the relationship between travel distance and travel probability.Finally,a recommendation list is generated by combining the effects of these two factors on a particular user.Compared with the state-of-the-art next new POI recommendation approaches,the experimental results demonstrate that STDMP achieves much better performance.