Research On Mobile Tourism Route Recommendation Model Based On Optimized Social Tags And Association Rules Algorithm

With the rapid development of mobile information technology and resources,information overload problem and information trek problem are becoming more and more serious.It is a hot spot that how to solve the mad complicated mobile information overload problem in mobile environment,which need to be solved in current mobile personalized recommendation field.As an effective solution to this problem,mobile recommendation system has received extensive attention and application from academics and mobile service providers.However,with the new features and new data of mobile recommendation,traditional recommendation systems are difficulted to apply directly in mobile recommendation field,and the existing mobile recommendation research is mostly based on the traditional recommender algorithm and not deeply or thorough.With the development of mobile tourism and personalized self-help travel,mobile travel recommendation of mobile travel platform has great value.However,with the rare core of the research and application in mobile travel route recommendation field,and most mobile tourism platform only provides popularity POI recommendation.To solve this problem,this paper constructs the overall architecture model of mobile recommendation system,and proposes two optimization schemes of UPST-TB and T-ARC for personalized recommendation algorithm analysis.Firstly,constructed the overall architecture model of the mobile recommendation system based on the intelligent network and clarified the overall architecture and described the specific process of the mobile recommendation system.The recommendation subsystem is used to implement the mobile personalized recommendation service,and the feedback subsystem is used to interact with users in real time to mining the real-time preferences and selection results of the user by using the iteration method.Secondly,combined with the mature semantic analysis technology and social tags technology in the Web3.0 era to solve the multi-dimensional mobile data of multi-data sources.And transformed important mobile data such as geographical location related information and mobile context into a single interest tags set.Then,combined it with user preferences and project characteristics,and established a "mobile users-social tagsprojects" three-part relationship network.By doing this,establish an UPST-TB mobile recommendation system.Which calculates the intrinsic connection by deep mining of interest tags and multiple tags weighted preprocessing methods to solve the problem of mobile information overload and improve the accuracy of recommendation.Furthermore,to solve the problems as cold start problem in mobile travel routes recommendation area.By optimizing association rule clustering recommender method and applied it in T-ARC mobile travel route recommendation algorithm.The process is that by association rules mining technology to extract correlation rules and avoid data sparseness such as missing user information;And then,representative high-level association rules are obtained to avoid redundancy and massive association rules for data mining according to the multi-layer association rule concept hierarchical tree;As the initial clustering applied to the k-means clustering algorithm,the user clustering method is recommended to solve the cold start problem.Finally,achieved mobile tourist route recommendation and greatly circumvented many recommender problems.Finally,the effectiveness of the UPST-TB algorithm and the T-ARC algorithm is proved by comparing the real data set with the traditional recommendation method and the improved algorithm.The results show that the deep mining of interest tags can improve the accuracy of recommendation.The association rule clustering algorithm can adapt to the recommendation field of mobile travel routes,and the optimization algorithm of high-level association rules is effective.