Landslide susceptibility mapping(LSM)based on different boosting and hyperparameter optimization algorithms:A case of Wanzhou District,China

作     者：Deliang Sun Jing Wang Haijia Wen YueKai Ding Changlin Mi 

作者机构：Key Laboratory of GIS Application ResearchChongqing Normal UniversityChongqing 401331China Key Laboratory of New Technology for Construction of Cities in Mountain AreaMinistry of EducationChongqing UniversityChongqing 400045China National Joint Engineering Research Center of Geohazards Prevention in the Reservoir AreasChongqing UniversityChongqing 400045China School of Civil EngineeringChongqing UniversityChongqing 400045China Natural Resources Development Service Center of LinyiLinyi 276000China 

出 版 物：《Journal of Rock Mechanics and Geotechnical Engineering》 (岩石力学与岩土工程学报（英文版）)

年 卷 期：2024年第16卷第8期

页      面：3221-3232页

核心收录：

学科分类：081504[工学-水利水电工程] 08[工学] 0815[工学-水利工程] 

基　　金：funded by the Natural Science Foundation of Chongqing(Grants No.CSTB2022NSCQ-MSX0594) the Humanities and Social Sciences Research Project of the Ministry of Education(Grants No.16YJCZH061) 

主　　题：Landslide susceptibility Hyperparameter optimization Boosting algorithms SHapley additive exPlanations(SHAP) 

摘      要：Boosting algorithms have been widely utilized in the development of landslide susceptibility mapping(LSM)***,these algorithms possess distinct computational strategies and hyperparameters,making it challenging to propose an ideal LSM *** investigate the impact of different boosting algorithms and hyperparameter optimization algorithms on LSM,this study constructed a geospatial database comprising 12 conditioning factors,such as elevation,stratum,and annual average *** XGBoost(XGB),LightGBM(LGBM),and CatBoost(CB)algorithms were employed to construct the LSM ***,the Bayesian optimization(BO),particle swarm optimization(PSO),and Hyperband optimization(HO)algorithms were applied to optimizing the LSM *** boosting algorithms exhibited varying performances,with CB demonstrating the highest precision,followed by LGBM,and XGB showing poorer ***,the hyperparameter optimization algorithms displayed different performances,with HO outperforming PSO and BO showing poorer *** HO-CB model achieved the highest precision,boasting an accuracy of 0.764,an F1-score of 0.777,an area under the curve(AUC)value of 0.837 for the training set,and an AUC value of 0.863 for the test *** model was interpreted using SHapley Additive exPlanations(SHAP),revealing that slope,curvature,topographic wetness index(TWI),degree of relief,and elevation significantly influenced landslides in the study *** study offers a scientific reference for LSM and disaster prevention *** study examines the utilization of various boosting algorithms and hyperparameter optimization algorithms in Wanzhou *** proposes the HO-CB-SHAP framework as an effective approach to accurately forecast landslide disasters and interpret LSM ***,limitations exist concerning the generalizability of the model and the data processing,which require further exploration in subsequent studies.