Resarch And Implementation On Personalized LBSN Travel Recommendations Algorithm

With the popularity of smart mobile devices, people can more easily use the Internet to provide services. In recent years, the rapid development of smart mobile device positioning technology and the widespread popularity of social networks, mobile Internet based on real-world location to provide users with personalized real-time services.More and more people use the Internet every day, a variety of information in the network, information overload problems with the emergence of people from the Internet to get the desired information more difficult. In order to solve the overload problem, so that users get their own sense of Interest information, the system in this article describes is by the analysis of the user's comments information and then get the user's favorite information for the user may be interested in the content and the user specified travel plans.In order to analyze the emotional tendencies expressed by the users in the text comments, this paper first proposes and establishes two emotion-oriented analysis models based on the emotional dictionaries. By comparing the model analysis results, the analytical model based on the AFINN emotional dictionaries is established.In order to explore the deep value of user reviews in social network,and use the AFINN dictionary to analyze the emotional tendencies of users' comments on Foursquare's sign-in and comment data with geographical location information, the user's implicit score for this product was obtained. Using the score as the basis,this paper proposes and implements an integrated user-based collaborative filtering recommendation algorithm, a geo-location-based recommendation, a user-friend recommendation hybrid recommendation algorithm, and recommends the content that he may be interested in. Then, Hybrid recommendation algorithm and single recommendation algorithm, the hybrid recommendation error is reduced to a certain extent and the data sparsity problem is solved.In order to further explore the value of recommending focused projects, this paper proposes a model of trip planning based on the hybrid recommendation model and travel frequent pattern set. This model can combine trip travel patterns with user preferences, Finally, the efficiency of the algorithm is tested and the feasibility of the itinerary is tested. The results of the test show that the proposed trip-making model can provide a feasible solution for the travel user to meet the driverse needs of the travel users. Personalized and reasonable trip planning table.