Crime risk factor analysis method considering spatial heterogeneity and excessive discrete phenomenon

Abstract

The invention discloses a crime risk factor analysis method considering spatial heterogeneity and an excessive discrete phenomenon. The method comprises the following steps of: taking a police stationarea as an analysis unit and taking the quantity of burglary cases as a modeling object, constructing a geographically weighted negative binomial model according to the geographic coordinates of thecentroid of the police station, the number of households and the environmental factor data, and alternately estimating the parameters of the geographically weighted negative binomial model by adoptingan improved iterative weighted least square method and a Newton-Raphson algorithm until the parameters converge; determining the optimal bandwidth of the geographically weighted negative binomial model by adopting an Akaike information criterion and cross validation, and calibrating the geographically weighted negative binomial model according to the optimal bandwidth; and estimating a regressioncoefficient of each environmental factor by adopting the calibrated geographically weighted negative binomial model, and obtaining the correlation between each environmental factor and the criminal activity according to the regression coefficient. According to the embodiment of the invention, the correlation between crime risk factors and crime activities can be evaluated more accurately.

Description

technical field [0001] The invention relates to the technical field of crime activity prediction, in particular to a crime risk factor analysis method taking into account spatial heterogeneity and over-dispersion phenomena. Background technique [0002] Due to the differences in the built environment and social demographic factors, the spatial distribution of criminal activities is often uneven and spatially clustered. In theory, the theory of social disorder is often used to explain the relationship between criminal activities and the social neighborhood environment, while the theory of daily activities and the theory of crime patterns are used to explain the spatial distribution of criminal activities. Technically, analytical models such as non-spatial regression models and spatial regression models can be used to evaluate the impact of crime risk factors on criminal activities. Through the effective judgment of the regression coefficient of the influencing factors of cri...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a crime risk factor analysis method that takes into account spatial heterogeneity and over-discrete phenomena, to solve the technical problem of inaccurate evaluation of the correlation between crime risk factors and criminal activities in the prior art, and to more accurately Assessing the correlation between crime risk and criminal activity

Images