An Efficient User-Friendly Integration Tool for Landslide Susceptibility Mapping Based on Support Vector Machines: SVM-LSM Toolbox

Landslide susceptibility mapping (LSM) is an important element of landslide risk assessment, but the process often needs to span multiple platforms and operation complex. This paper develops efficient user-friendly toolbox including whole LSM, known as SVM-LSM toolbox. The realizes based on a support vector machine (SVM), which can be integrated into ArcGIS or Pro platform. includes three sub-toolboxes, namely: (1) influence factor production, (2) selection dataset (3) model training prediction. Influence production provides automatic calculation DEM-related topographic factors, converts line data continuous raster performs rainfall processing. Factor uses Pearson correlation coefficient (PCC) calculate correlations between information gain ratio (IGR) contributions different factors occurrence. Dataset sample generation non-landslide data, split. accuracy, precision, recall, F1 value, receiver operating characteristic (ROC) area under curve (AUC) are used evaluate prediction ability model. In addition, two methods—single processing multiprocessing—are generate LSM. efficiency multiprocessing much higher than that single process. order verify performance accuracy toolbox, Wuqi County, Yan’an City, Shaanxi Province was selected test results show AUC value 0.8107. At same time, tool improves by about 60%. experimental confirm practicability proposed in