Geologic hazard risk assessment based on machine learning
For the problems of inaccurate factor coefficients and unreasonable risk zoning in geologic haz-ard risk assessment,this paper proposes a machine learning model geologic hazard risk assessment meth-od based on grid from the perspectives of multi-factor and multi-feature of geologic hazard.Firstly,it collects and organizes the geohazard data in Wenzhou City,uses the spatial analysis and attribute calcula-tion function of ArcGIS,introduces the neural network model to optimize and screen the weighting fac-tors,and takes the selected geomorphic type,topographic slope,road construction,geological structure,soil type,river distribution,vegetation coverage as the geologic hazard evaluation factors,and gets the weighting coefficients of each preferred factor.Finally,it uses the weighted analysis results in the evalua-tion map of geologic hazard risk factors.The results show that the automatic calculation of evaluation fac-tors and weight coefficients by the machine learning model of subdivided grid cells can improve the ration-ality of the evaluation of geohazard risk,which can provide a scientific basis and reference for the man-agement of regional geohazard prevention and mitigation.
geologic hazardrisk assessmentGIS spatial analysisneural network model
万方数据
维普
