Hie-DensityR: A Model For Extracting Boundaries And Spatial Structures Of Tourist Check-in Activity Zones

The spatial activities of tourists in tourist destinations follow potential patterns and rules,but there are also significant differences between different types of destinations.Quantitative analysis of the spatial activity distribution hotspots and spatial structure of tourists in the destination helps to discover hidden characteristics of tourist behavior and activity spatial distribution,understand the general and regional rules of tourist spatial activity,promote the rational and efficient allocation of tourism resources,and the construction and management of smart tourism.Based on the Hie-Density check-in activity boundary extraction method,this article aims to extract the multi-scale regional range of tourist check-in activity spatial boundaries and analyze the quantitative structure of check-in activity hotspots.A Hie-Density R model is constructed and empirical research is carried out using Flickr check-in data from inbound tourists in Beijing.The main achievements and conclusions of this paper are as follows:(1)The Hie-Density check-in activity boundary extraction method is improved,and the Hie-Density R model is developed to achieve region-based check-in activity spatial boundary extraction and quantitative calculation of spatial structure.The basic idea of the model is as follows: first,a Delaunay triangulation network based on check-in points is generated to form a global check-in activity network.Weakly connected edges are removed to obtain a surface that can represent the adjacent network of tourist check-ins,which is then transformed into a check-in activity surface using kernel density analysis.Second,the check-in activity surface is partitioned into continuous ring-layer structures using contour lines.The theoretical radius of each level of the ring layer and the number of check-in points are plotted as a ring-layer expansion curve to reflect the expansion characteristics of check-in activity from the aggregation center to the edge,and the tourist activity aggregation coefficient is calculated.The Hie-Density method is optimized to calculate the spatial utilization of each level of the ring layer and obtain the optimal boundary of the check-in activity space.Finally,the check-in hotspots contained within the optimal boundary are sorted by area,and their positional size characteristics are analyzed.The spatial structure characteristics of each check-in hotspot are quantitatively expressed using the local contour line tree construction method,and different check-in hotspots are classified into three types of structures: single branch,main and secondary branch,and multi-branch,demonstrating the characteristics and evolutionary process of checkin activity space.(2)A fully automated calculation framework for the Hie-Density R model was designed based on Python scripts.Using open-source geospatial processing packages such as Geo Pandas,Shapely,and Arc Py library of Arc GIS,a quantitative calculation program was established for automatically extracting the spatial boundaries of check-in activities and calculating the spatial structure of check-in activity hotspots,starting from check-in data input and initial parameter settings.In terms of model implementation and calculation process,methods such as optimizing the algorithm for constructing the network of nearby check-ins and establishing spatial indexes greatly improved the computational efficiency of the model,making it suitable for studying tourist check-in activities in multi-scale big data scenarios.(3)Using Flickr check-in data from inbound tourists to Beijing,the optimal boundaries of check-in activity spaces were extracted and analyzed,and the spatial structure characteristics of check-in hotspots within the boundaries were analyzed.This study used the Hie-Density R model to extract and compare the ring expansion features,optimal boundary shape,distribution of check-in hotspot sequence size and spatial structure features,and evolution rules of checkin data in different time spans(yearly and seasonally),in order to investigate the spatial distribution and evolution of inbound tourists at different stages and periods.Finally,this study compared the differences in the spatial structure of check-in activity boundaries and hotspots between inbound tourists in Beijing and other regions such as Shanghai and Guangzhou.The Hie-Density R model is suitable for extracting the boundaries of tourist check-in activity spaces in multi-scale regional ranges and can effectively identify the clustering and diffusion characteristics of check-in data spatial distribution,objectively and accurately expressing the spatial structure of tourist check-in activity boundaries and hotspots.By analyzing the check-in activity spaces of inbound tourists in Beijing,this model can quantitatively identify the optimal boundaries of check-in activities,the spatial distribution of check-in hotspots,the sequence size distribution,and spatial structure features,providing a feasible reference for geographic and temporal data mining research and theoretical and technical support for scientific and reasonable tourism planning.