Personalized POI Recommendation Based On User Preferences

With the rapid development of information technology and the popularization of intelligent mobile terminals,the era of information everywhere is coming.Personalized recommendation system is one of the common methods to solve the problem of information overload.Using personalized recommendations not only helps users find the information they need more quickly,but also helps merchants to display the information they generate more quickly and effectively in front of consumers.The POI recommendation enables users to share real-time geographic information in real time through check-in.Users will be able to rate and comment on the items on the check-in site through mobile devices,record their feelings about the project,and also see their friends' check-ins and interact with them.In this paper,POI recommendation algorithm is studied in depth,and the commonly used personalization algorithm is introduced.At the same time,the common POI recommendation model.In view of the problems existing in the personalized POI recommendation algorithm this paper has made the improvement,mainly includes four parts:(1)the time and location attributes are based on the geographic location of the web services in the important attribute,this paper puts forward in the process of computing similarity based on the weight of time and location factors regulating method;(2)introduce the Ebbinghaus forgetting curve,propose the nonlinear forgetting function and explore the change of parameters;(3)the Slope One algorithm is introduced into POI recommendation system,the improved algorithm commonly used compared advantages and disadvantages,puts forward the improved nearest neighbor Slope One algorithm based on user preferences.(4)the performance and the advantages and disadvantages of the algorithm are analyzed based on the real data.