Landslide hazard zonation using LS-SVM and GIS analytic functions

Landslide hazard zonation using LS-SVM and GIS analytic functions

Yousef Ebadi1

1) MS.c student in RS and GIS – Tabriz University- Tabriz – Iran

IOI Code : XBAE-DZFGZ
Publication : Demo - Website of Saman Conference Management System(ngtu.ir)
Abstract :
    Landslide, as one of the natural behaviors of the environment, has always caused great damage to human construction. Which is sometimes overlooked and not paid much attention by the planners. The purpose of this study was to evaluate the accuracy of one of the computational models of artificial intelligence (least squares support vector machine), which is a modified and upgraded model of the SVM classic model. 9 significant and important factors affecting landslide were identified, and normalization was based on a relatively high frequency. of the 109 recorded slides points, 70% were used for training the model and 30% for the validation of the made model. The model validation results using RMSE-ROC-AUC criteria indicate the high accuracy of this method in predicting domain instability. And the results are also applicable to planners and custodians

Keywords : landslide soft computing ls svm validation curve