Analysis Of Temporal And Spatial Characteristics Of Tourist Flow In Scenic Spots Based On Tencent Location Big Data

The rapid development of social networking and location-based services has changed the travel behavior of tourists.Tourists can use social media to obtain travel information or share travel experiences anytime and anywhere before,during,and after travel.These behaviors generate rich spatio-temporal data information,which more effectively and accurately records the spatio-temporal behaviors of tourists in tourism activities,and provides a new data source for the spatio-temporal research of tourist flow.Tencent’s location big data platform is based on this platform to provide massive location data analysis and technical services.The regional thermal map reflects the temporal and spatial distribution and changes of tourist flow in some scenic spots,cities,transportation distribution centers,large shopping places and other regions under different time granularity according to the positioning data.Compared with other location big data,Tencent’s regional heat map data has the advantages of real-time,wide coverage of user groups,and high spatial positioning accuracy.Therefore,this paper selects Tencent regional thermal map as the basic data for the study of short-termtourist flow in scenic spots.Taking Beijing Happy Valley Scenic Spot as a case,this paper analyzes the temporal and spatial variation of tourist flow in the scenic spot by using mathematical statistics,kernel density estimating and density-field based hotspot detector.Then,according to the fluctuation and changing trend of the tourist flow sequence itself,a short-term forecasting model of tourist flow is constructed.The main research conclusions are as follows:1.Analyzes the temporal distribution characteristics of tourist flow in Beijing Happy Valley scenic spot from three aspects: holidays,weeks and days.During the holidays,the tourist flow shows an "inverted V" pattern,and the peak occurs on the second day of the holiday;the intra-week variation curve presents an obvious "lifting tail" shape,and Saturday and Sunday are the peak tourist flow days;from the perspective of the daily time scale,the change trend of tourist flow on weekdays and weekends is consistent,the tourist flow curve presents an "inverted U" shape,the tourist flow is on the rise from 8:00 to 14:00,the peak tourist flow is formed from 14:00 to 18:00,and shows a downward trend from 18:00 to 22:00 at night.2.Using kernel density analysis method and density field hot spot detection model,this paper analyzes the overall spatial distribution characteristics and intra day spatial dynamic change law of tourist flow.On the whole,the spatial distribution shows the characteristics of "multi-point gathering",with four obvious hot spots,which show the characteristics of gathering to the main tourism projects in the scenic spot;The spatial variation law of intra day tourist flow can be roughly divided into three periods: 8:00-10:00 is the gathering period,10:00-19:00 is the peak period,and19:00-22:00 is the dispersion period.3.Based on the historical tourist flow itself,the SARIMA prediction model and the seasonal index adjustment-SARIMA forecast model are constructed and the prediction effect of the model is evaluated using RMSE and MAPE.It is found that the seasonal index adjustment-SARIMA has better prediction effect.Finally,the suggestions about early warning and regulation of tourist flow are proposed from three aspects: regulation of low-peak period,regulation of intraday tourist flow,and regulation of spatial tourist flow.