Research And Empirical Study On Short Time Tourism Route Planning Method Based On Multi-source Data

With the vigorous development of China's tourism industry and the continuous improvement of people's living standards,free travel has gradually become the first choice for people to travel.With the rapid development of the Internet,the number of tourism information resources is increasing.It is difficult for people to formulate their own travel plans through information retrieval.It often takes a lot of time and energy before traveling,so the need for automated travel route planning is also With the development of tourism,it has increased.This article takes the short-time experiential tourism as the starting point,and uses the multi-source data with geographical location to collect useful tourist information by using taxi trajectory data,traffic data,and travel route data uploaded by users spontaneously on the Internet,and considering the spatiotemporal relationship of scenic spots.Degree and transit time,combined with relevant models and algorithms,designed a short-term travel route planning method for the characteristics of experiential travel time constraints,planning and developing the optimal travel route for passengers to meet their time requirements.Realize a shorttime travel route planning platform with user interaction mode to realize visual application.The research work of this paper is as follows:Using taxi trajectory data,traffic data,and user-sponsored travel route data to mine useful travel information,adding time constraints to traditional route planning methods,enabling route planning by hour,not only expanding data The scope of application also meets the needs of user experience travel.Search for tourist routes by pre-calculation.Using the idea of graph theory,the scenic spot data is mapped into a directed weighted map,with nodes representing the scenic spots,and the directed edges representing the feasible routes between the two scenic spots.The weight of the side represents the time spent visiting two scenic spots,including the time of scenic spots and the transit time between two scenic spots.Combined with the popularity of visiting the scenic spot and the degree of association between the two scenic spots,a breadth-first traversal-based strategy is used to search for all feasible routes that meet the time constraints.Using the regression model to score and sort the travel routes,the purpose is to identify the sequence of route output more quickly during the calculation.Using the travel route data as the test data set,using the mean square error,the interpretation variance,and the R square score as the test indicators,five regression models were selected for the test,and the comparison test index and the model calculation time were comprehensively selected.Finally,the mean square error was selected and the variance was interpreted.And the random forest regression instrument with the highest R score is used for the short-term travel route scoring model.Develop a short-time travel route planning platform using B/S mode,complete the seamless connection between the terminal and the data server by introducing a Web server,adopt C#,ASP.NET,AJAX terminal development technology,and use stable and efficient background database management.Oracle implementation.The input and output process is realized through the user interaction experience mode,and the running result is displayed on the map in combination with the Gaode map API,which intuitively tells the user which routes can be selected,and displays the tour order of each route.