Application of Geographical Weighted Regression Model inAnalyzing Internet Fraud
YAN Wenfeng
SUN Lingli
Abstract：By using the data of Internet fraud research report of China in 2014,collecting and collating the national 31 provincial scale regions,and combining with relevant information,we finally select the proportion of urban population,network penetration rates,GDP of all provinces,the total population for each province,and the proportion of men as explanatory variables.After exploring the spatial variation of the number of Internet fraud and its influencing factors,we have found that the numbers of Internet fraud for 31 provinces have a spatial autocorrelation,which shows that geographical weighted regression model can be applicable.By comparing GWR model with ordinary linear regression model,we find that GWR model is superior to OLR model and GWR model can deeply reveal the spatial relationship between the numbers of Internet fraud and the influencing factors.
Keywords：internet fraudspatial autocorrelationgeographic weighted regression
Publication Date：2017-01-01
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：4( 148-151 )