Impact of Non-Landslide Sample Sampling Strategies and Model Selection on Landslide Susceptibility Mapping

This study investigated the influence of non-landslide sampling strategies on landslide susceptibility assessment (LSA) performance and explored approaches to minimizing uncertainty in model selection. Five non-landslide sampling strategies were evaluated using the random forest (RF) model to genera...

Full description

Saved in:
Bibliographic Details
Main Authors: Weijun Jiang, Ling Li, Ruiqing Niu
Format: Article
Language:English
Published: MDPI AG 2025-02-01
Series:Applied Sciences
Subjects:
Online Access:https://www.mdpi.com/2076-3417/15/4/2132
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850082586590707712
author Weijun Jiang
Ling Li
Ruiqing Niu
author_facet Weijun Jiang
Ling Li
Ruiqing Niu
author_sort Weijun Jiang
collection DOAJ
description This study investigated the influence of non-landslide sampling strategies on landslide susceptibility assessment (LSA) performance and explored approaches to minimizing uncertainty in model selection. Five non-landslide sampling strategies were evaluated using the random forest (RF) model to generate landslide susceptibility maps (LSMs) for each scenario. To assess the impact of these strategies, this study employed a receiver operating characteristic (ROC) curve, a confusion matrix, and various statistical indicators. Additionally, the mean susceptibility indices derived from the gradient boosting decision tree (GBDT), support vector machine (SVM), and RF models were analyzed to evaluate their effectiveness in reducing the uncertainty during model selection. The GBDT, SVM, and RF were selected for their ability to handle complex, nonlinear relationships in the data, superior generalization capability, effective mitigation of overfitting risks, high predictive performance, and robustness. The findings revealed that selecting non-landslide samples from slope units without landslides enhances accuracy and averaging across models mitigated the uncertainty associated with landslide susceptibility models. Furthermore, this study demonstrated that the non-landslide sample selection method significantly improved prediction accuracy, particularly when samples were drawn from very-low-susceptibility zones identified by pre-classified machine learning models. These results highlight the importance of refining sample selection strategies and integrating multiple machine learning models to improve the reliability and accuracy of landslide susceptibility assessments. This approach provides valuable insights for future research and practical applications in risk mitigation and disaster management by offering a more precise depiction of low-susceptibility areas, thereby reducing the occurrence of false positives in landslide prediction.
format Article
id doaj-art-ab0e55c49a734cd39b74a7015fe51da0
institution DOAJ
issn 2076-3417
language English
publishDate 2025-02-01
publisher MDPI AG
record_format Article
series Applied Sciences
spelling doaj-art-ab0e55c49a734cd39b74a7015fe51da02025-08-20T02:44:29ZengMDPI AGApplied Sciences2076-34172025-02-01154213210.3390/app15042132Impact of Non-Landslide Sample Sampling Strategies and Model Selection on Landslide Susceptibility MappingWeijun Jiang0Ling Li1Ruiqing Niu2School of Future Technology, China University of Geosciences, Wuhan 430074, ChinaSchool of Future Technology, China University of Geosciences, Wuhan 430074, ChinaSchool of Future Technology, China University of Geosciences, Wuhan 430074, ChinaThis study investigated the influence of non-landslide sampling strategies on landslide susceptibility assessment (LSA) performance and explored approaches to minimizing uncertainty in model selection. Five non-landslide sampling strategies were evaluated using the random forest (RF) model to generate landslide susceptibility maps (LSMs) for each scenario. To assess the impact of these strategies, this study employed a receiver operating characteristic (ROC) curve, a confusion matrix, and various statistical indicators. Additionally, the mean susceptibility indices derived from the gradient boosting decision tree (GBDT), support vector machine (SVM), and RF models were analyzed to evaluate their effectiveness in reducing the uncertainty during model selection. The GBDT, SVM, and RF were selected for their ability to handle complex, nonlinear relationships in the data, superior generalization capability, effective mitigation of overfitting risks, high predictive performance, and robustness. The findings revealed that selecting non-landslide samples from slope units without landslides enhances accuracy and averaging across models mitigated the uncertainty associated with landslide susceptibility models. Furthermore, this study demonstrated that the non-landslide sample selection method significantly improved prediction accuracy, particularly when samples were drawn from very-low-susceptibility zones identified by pre-classified machine learning models. These results highlight the importance of refining sample selection strategies and integrating multiple machine learning models to improve the reliability and accuracy of landslide susceptibility assessments. This approach provides valuable insights for future research and practical applications in risk mitigation and disaster management by offering a more precise depiction of low-susceptibility areas, thereby reducing the occurrence of false positives in landslide prediction.https://www.mdpi.com/2076-3417/15/4/2132landslide susceptibilitymachine learningmodeling uncertaintyremote sensing
spellingShingle Weijun Jiang
Ling Li
Ruiqing Niu
Impact of Non-Landslide Sample Sampling Strategies and Model Selection on Landslide Susceptibility Mapping
Applied Sciences
landslide susceptibility
machine learning
modeling uncertainty
remote sensing
title Impact of Non-Landslide Sample Sampling Strategies and Model Selection on Landslide Susceptibility Mapping
title_full Impact of Non-Landslide Sample Sampling Strategies and Model Selection on Landslide Susceptibility Mapping
title_fullStr Impact of Non-Landslide Sample Sampling Strategies and Model Selection on Landslide Susceptibility Mapping
title_full_unstemmed Impact of Non-Landslide Sample Sampling Strategies and Model Selection on Landslide Susceptibility Mapping
title_short Impact of Non-Landslide Sample Sampling Strategies and Model Selection on Landslide Susceptibility Mapping
title_sort impact of non landslide sample sampling strategies and model selection on landslide susceptibility mapping
topic landslide susceptibility
machine learning
modeling uncertainty
remote sensing
url https://www.mdpi.com/2076-3417/15/4/2132
work_keys_str_mv AT weijunjiang impactofnonlandslidesamplesamplingstrategiesandmodelselectiononlandslidesusceptibilitymapping
AT lingli impactofnonlandslidesamplesamplingstrategiesandmodelselectiononlandslidesusceptibilitymapping
AT ruiqingniu impactofnonlandslidesamplesamplingstrategiesandmodelselectiononlandslidesusceptibilitymapping