AUTHOR=Xia Liheng , Shen Jianglong , Zhang Tingyu , Dang Guangpu , Wang Tao 

TITLE=GIS-based landslide susceptibility modeling using data mining techniques

JOURNAL=Frontiers in Earth Science

VOLUME=Volume 11 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/earth-science/articles/10.3389/feart.2023.1187384

DOI=10.3389/feart.2023.1187384

ISSN=2296-6463

ABSTRACT=Landslide is a most widespread geohazards around the world. Therefore, it is necessary and meaningful to map regional land-slide susceptibility for landslide mitigation. In this research, land-slide susceptibility maps were produced by four models, name-ly, Certainty factors (CF), Naive Bayes (NB), J48 decision tree (J48), and Multilayer Perceptron (MLP) models. First step, 328 land-slides were identified via historical data, interpretation of remote sensing images and field investigation, and they were divided into two subsets which assigned different uses: 70% subset for training and 30% subset for validating. Then, twelve conditioning factors were employed, including altitude, slope angle, slope aspect, plan curvature, profile curvature, TWI, NDVI, distance to rivers, distance to roads, landuse, soil and lithology. Later, analyze the importance of each conditioning factor by average merit (AM) values, and evaluate the relationship between landslide occurrence and various factors using certainty factor (CF) approach. In the next step, the landslide susceptibility maps were produced based on four models, and quantitatively compare the effect of four models by receiver operating characteristic (ROC) curves, area under curve (AUC) values and nonparametric tests. The re-sults demonstrated that all the four models can reasonably assess landslide susceptibility. Of these four models, CF model has the best predictive performance for the training (AUC=0.901) and validating data (AUC=0.892). The proposed approach is an innova-tive method that may also help other scientists to develop landslide susceptibility maps in other areas but also as an approach that could be used in geo-environmental problems besides natural hazard assessments.