Abstract
The main purpose of this study was to evaluate and compare two data-mining techniques, functional data analysis (FDA) and the generalized linear model (GLM), for landslide susceptibility assessment. In total, 202 landslide locations were identified from interpretation of multitemporal satellite imagery and Google Earth data, and extensive field surveys, among these, 141 (70%) landslides were randomly selected as training data and the remaining 61 (30%) cases were used for validation. Twelve landslide conditioning factors were prepared, including slope aspect, digital altitude, distance to fault, land use, lithology, normalized difference vegetation index, plan curvature, distance to river, distance to road, slope angle, soil type, and topographic wetness index. Afterward, the landslide susceptibility maps were produced using FDA and GLM models in R statistical environment. Finally, the results of the models were validated and compared using the area under the curve (AUC), standard error (std. error), confidence interval (CI) at 95%, and significance level (P value). The FDA model, with an AUC value of 0.722, std. error of 0.048, CI of 0.627–0.816, and significance level P of .000, showed better performance than GLM in the present study.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: Spatial Modeling in GIS and R for Earth and Environmental Sciences
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
