Research On Correlation Between Tourism POI And Regional Economy Based On GWR

As the basic data of spatial big data,tourism POI data can reflect the spatial structure and spatial distribution of tourism resources related objects.The information recorded by the tourism POI point can also reflect the economic and social functions carried by the tourist attractions and related space entities to a certain extent.A certain range of spatial areas covers different quantities and categories of tourism POI data.The information such as the category,quantity and spatial location of these POI data can preliminarily reflect the spatial distribution pattern of the structure and volume of the tourism economy.Exploring and analyzing the correlation between tourism POI and regional economy and the spatial heterogeneity of its related strength can provide a certain reference for regional tourism resource allocation and overall development.The correlation between tourism POI and the regional economy is generally measured by the least squares regression method,and its regression coefficient is used to measure the correlation strength between the independent variable and the dependent variable.However,least squares regression is a global regression model.It can only describe the correlation between variables as a whole.In some applications,it ignores the spatial properties of variables,which will cause large deviations in the model fitting results.The local area is effectively fitted.The geographic weighted regression method based on local smoothing considers the spatial factor of the variable.The regression coefficient of the regression model independent variable is able to perform local area fitting as the spatial position of the variable changes.To this end,this paper uses geographic weighted regression to analyze the correlation between tourism POI and regional economy and its spatial heterogeneity.First,obtain tourism POI and related data,and perform pre-processing and analysis to prepare sample data for the regression model.Collect administrative divisions and GDP statistics of the study area,and transform and correlate the data;write web crawlers based on Gaode map API,obtain tourism-related POI data,and perform data cleaning and data storage;explore using nuclear density method The spatial distribution of different types of POI point data.Then,with the Chinese mainland as the research area,440 prefecture-level cities were selected as statistical units.Based on the various types of POI data corresponding to the six elements of tourism,the independent variables were constructed,and the regression model was constructed with the GDP of the tertiary industry as the dependent variable.Diagnostic analysis.The correlation coefficient,the variable scatter matrix,the eigenvalue and the conditional index method are used to test the collinearity of the independent variables,and the independent variables of the regression model are selected.The Moran's I index is used to analyze the spatial autocorrelation of the dependent variables to determine whether the dependent variable is Meet the prerequisites of the geographically weighted regression model.A regression model was constructed for the independent variables and dependent variables,and the least squares regression model and the geographically weighted regression model were used for modeling analysis.The fitting effects of the two models were compared and analyzed.Finally,the spatial heterogeneity of tourismrelated POI and regional economic correlation is analyzed.The GIS visualization method is used to graphically display the regression coefficients of each explanatory variable in the GWR model results,and analyze the spatial heterogeneity of tourism POI and regional economic correlation.The experimental results show that the fitting degree(R2)of the geographically weighted regression model is improved by 13.6% compared with the least squares regression model in the regression model of tourism POI and regional economy.The fitting effect of geographically weighted regression model Better than the least squares regression model.There is a certain correlation between tourism POI and regional economy,and its relevance has obvious spatial heterogeneity.The area with the highest correlation between scenic spot POI and regional economy appears in South China;the spatial difference of shopping service POI and regional economic correlation is not big;the correlation between sports leisure service POI and regional economy is more different in South China and East China.The spatial difference between the accommodation service POI and the regional economy is mainly concentrated in the southern region of North China and the northern region of East China.In summary,this paper builds a geographically weighted regression model based on tourism-related POI and regional economy,and explores the spatial distribution characteristics of tourism POI and regional economic relevance.It has certain practical significance in the distribution and planning management of tourism resources.