摘要
The main purpose of this study is to analyze and compare the results of spatial assessment of landslide hazards by means of logistic regression analysis,back-propagation artificial neural network (ANN) through the use of geographic information system (GIS) techniques.A region from the Three Gorges Dam to the Badong Town,which suffered from deformation for many years,was selected as the application site of this study.Firstly,nine evaluation factors were selected,including topographic slope,topographic aspect,soft interlayer,water system influence area,land use,and so on.Then,this paper highlighted the discrepancies between a logistic regression model and an ANN model,made a landslide susceptibility map,and evaluated the performance of these two models by ROC curve analysis. Finally,based on the above analyses,satisfactory agreement was obtained between the susceptibility map and the existing data on landslide location.The area under curve (AUC) values of the logistic regression and the ANN model are 0.806 and 0.799.Accuracies of these two models can be evaluated relatively similarly.