Design And Implementation Of Point Of Interest Recommendation System Based On Trajectory Data Analysis

With the popularization of smart terminals,and the continuous maturity of mobile positioning technology,a large amount of data that incorporates geographic location information has been generated,and trajectory data is one of them.Trajectory data can represent the mobility of objects that have the ability to move(such as humans,animals,and vehicles),and it has a high mining value in the geographic-based service field.At present,there have been many applications in the field of trajectory data mining,such as the recommendation of point of interest,that is,recommending a place that meets his wishes to the user.At the same time,the recommendation of point of interest has also emerged with the continuous development of location-based social networks Common applications.Point of interest recommendation not only allows users to find a place that meets their wishes in this era of data explosion,but also provides considerable economic benefits for businesses.Most traditional point of interest recommendation methods make recommendations based on the check-in data generated in the location social network,that is,inferring the user's preference for other places that have not checked in based on the user's historical check-in information.However,the generation of check-in data depends on the user's willingness to check in to the place,which is sparse.The trajectory data of daily travel of users can objectively display the behavioral rules of users,exposing users' preferences for places they have visited in the past.Therefore,this article will design and implement a recommendation system based on trajectory data,and analyze historical points of interest based on user trajectory data before making recommendations.First,the trajectory data preprocessing,stay point extraction,stay point clustering and other processing algorithms for trajectory data are introduced,as well as the existing recommendation algorithms.The algorithm is then validated using the trajectory dataset.Finally,the system requirements of the point of interest recommendation system based on trajectory data analysis are described,and detailed functional modules are divided according to the system requirements to complete the system architecture design,database design and interface design,and the system is implemented using the SSM framework.The paper uses the precision rate and recall rate to evaluate the recommendation algorithm in the algorithm research stage,and the experiment proves that the recommendation effect meets the system requirements.Finally,a functional test was performed on the point of interest recommendation system based on trajectory data analysis,which verified the function of the system and showed that the system can be used normally.