Research And Implementation Of Successive Location Recommender System In Shopping Area

Location recommender system is the typical application of location based service. When users are unable to quickly and accurately find the location in line with their own interests in a particular area for their limited of this area information, location recommendation system can provide effective help for the user. Users cann’t quickly find the locations to meet their own interest in shopping area for the diversity of location category and the comprehensive services, so the shopping area becomes a good choice for location recommendation system deployment. The recommendation system implemented by this paper can provide the service described as follows: in particular shopping area, based on the user’s current location and historical information, providing successive location recommendation and helping the user get real-time information of recommended locaitons to make dicision.This paper firstly analyzed the shopping area and user’s particular scenario. Based on the particular scenario and users’ history path, proposed the concept of hierarchical path. Then by the similar path of user hierarchical path, computed the similarity of users. At the same time, based on the history path, computed the location-location Markov transition matrix for every user. Then got the expected probabilities of a user from the current location to the other locations by matrix operations, selected the locations have the highest probability to recommend. Meanwhile, this paper also designed a method to get real-time information of recommended locations: the system interacted with the location-based social network to access the check-in users of social network at recommended locations, and then selected the most likely to answer problem candidates by filtering and sorting, asked them question about the real-time information of recommended locations, and got their answer as the real-time information. On the basis of the above methods, this paper designed and implemented successive location recommendation system in shopping area, using offline computing to get similarity matrix and Markov transition matrix, using online computing to handle user’s request and get recommended locations by matrix operations, using client to show the result and help user get real-time information of these locations. At last, this paper proved this recommendation method based on the actual data.This paper firstly introduced the related research background, and reviewed the existing problems of related research. Then analyzed the shopping area and user’s particular scenario, proposed the concept of hierarchical path. And made a successive location recommendation method of shopping area, meanwhile designed a method to get real-time information of recommended locations. At last, this paper described the implementation of such system, and proved this recommendation method based on the actual data.