Evaluation of Geological Disaster Sensitivity in Shuicheng District Based on the WOE-RF Model

To improve the prevention and control of geological disasters in Shuicheng District, 10 environmental factors—slope, slope direction, curvature, NDVI, stratum lithology, distance from fault, river system, annual average rainfall, road land use—were selected as evaluation indicators by integrating factors such landform, basic geology, hydrometeorology engineering activities. Based on weight evidence, random forest, support vector machine BP neural network algorithms were introduced to build WOE-RF, WOE-SVM WOE-BPNN models. The sensitivity District was evaluated using GIS platform, region divided into areas extremely high, medium, low disasters. By comparing analyzing ROC curve distribution law index, AUC accuracy models 0.836, 0.807 0.753, respectively; WOE-RF model shown be most effective. In model, high-, medium-, low- low-sensitivity accounted for 15.9%, 16.9%, 19.3%, 21.0% 26.9% study area, respectively. high- high-sensitivity are mainly concentrated with large slopes, broken rock masses, systems intensive human activity. These research results consistent actual situation can provide a reference this similar mountainous areas.