Landslide risk assessment and management using hybrid machine learning‐based empirical models DOI Open Access
Dingying Yang, Xi Jiang, Alireza Arabameri

et al.

Geological Journal, Journal Year: 2023, Volume and Issue: 59(3), P. 885 - 905

Published: Oct. 30, 2023

Landslides are a prevalent geologic phenomenon that substantially threatens human life and infrastructure, resulting in considerable loss destruction. The practice of landslide susceptibility mapping is crucial for the mitigation risks connected with this natural disaster. This work aims at investigating influence varying sample sizes on precision modelling using case study conducted Alamout basin, Iran. researchers used machine learning methodology based tree algorithms to construct model predicting likelihood landslides. Additionally, they adopted multi‐scenario strategy address inherent uncertainty associated input data. integration naive Bayes (NBTree), random forest (RF), logistic (LMT) J48 was performed. process included 20 predictive parameters across four distinct scenarios. Four models, labelled S1, S2, S3 S4, were study. These models utilized 25%, 50%, 75% 100% available inventory research presented distinguished by tree‐based incorporating findings indicated augmentation size improved models. efficacy enhancing dependability also underscored. Among elements process, it seen slope angle accounted highest relative significance, constituting 25.60% overall influence. Following more closely, distance fault contributed significantly, importance 23.40%. rainfall elevation exhibited notable contributions, volumes 7.91% 5.50%, respectively. All showed adequate forecasting ability throughout training testing phases. During phase, true skill score (TSS) values range 0.631–0.804, while area under receiver operating characteristic curve 0.745–0.921. maps significant portion region exhibits moderate very high zones, northern eastern sectors displaying greater than western region. model's performance improvement from S1 S4 both following trend: scenario 1, RF outperformed J48, LMT NBTree models; 2, surpassed being par model; scenarios 3 4, superior compared NBTree, Therefore, proved be most effective among evaluated. derived have potential serve as valuable references purposes land‐use planning catastrophe risk management.

Language: Английский

Automated Machine Learning-Based Landslide Susceptibility Mapping for the Three Gorges Reservoir Area, China DOI
Junwei Ma,

Dongze Lei,

Zhiyuan Ren

et al.

Mathematical Geosciences, Journal Year: 2023, Volume and Issue: 56(5), P. 975 - 1010

Published: Nov. 28, 2023

Language: Английский

Citations

36
Assessment of the effects of characterization methods selection on the landslide susceptibility: a comparison between logistic regression (LR), naive bayes (NB) and radial basis function network (RBF Network) DOI
Hui Shang,

Lixiang Su,

Yang Liu

et al.

Bulletin of Engineering Geology and the Environment, Journal Year: 2025, Volume and Issue: 84(3)

Published: Feb. 15, 2025

Language: Английский

Citations

1
Assessment of land degradation susceptibility within the Shaqlawa subregion of Northern Iraq-Kurdistan Region via synergistic application of remotely acquired datasets and advanced predictive models DOI

Badeea Abdi,

Kamal Kolo, Himan Shahabi

et al.

Environmental Monitoring and Assessment, Journal Year: 2024, Volume and Issue: 196(11)

Published: Oct. 25, 2024

Language: Английский

Citations

1
Landslide risk assessment and management using hybrid machine learning‐based empirical models DOI Open Access
Dingying Yang, Xi Jiang, Alireza Arabameri

et al.

Geological Journal, Journal Year: 2023, Volume and Issue: 59(3), P. 885 - 905

Published: Oct. 30, 2023

Landslides are a prevalent geologic phenomenon that substantially threatens human life and infrastructure, resulting in considerable loss destruction. The practice of landslide susceptibility mapping is crucial for the mitigation risks connected with this natural disaster. This work aims at investigating influence varying sample sizes on precision modelling using case study conducted Alamout basin, Iran. researchers used machine learning methodology based tree algorithms to construct model predicting likelihood landslides. Additionally, they adopted multi‐scenario strategy address inherent uncertainty associated input data. integration naive Bayes (NBTree), random forest (RF), logistic (LMT) J48 was performed. process included 20 predictive parameters across four distinct scenarios. Four models, labelled S1, S2, S3 S4, were study. These models utilized 25%, 50%, 75% 100% available inventory research presented distinguished by tree‐based incorporating findings indicated augmentation size improved models. efficacy enhancing dependability also underscored. Among elements process, it seen slope angle accounted highest relative significance, constituting 25.60% overall influence. Following more closely, distance fault contributed significantly, importance 23.40%. rainfall elevation exhibited notable contributions, volumes 7.91% 5.50%, respectively. All showed adequate forecasting ability throughout training testing phases. During phase, true skill score (TSS) values range 0.631–0.804, while area under receiver operating characteristic curve 0.745–0.921. maps significant portion region exhibits moderate very high zones, northern eastern sectors displaying greater than western region. model's performance improvement from S1 S4 both following trend: scenario 1, RF outperformed J48, LMT NBTree models; 2, surpassed being par model; scenarios 3 4, superior compared NBTree, Therefore, proved be most effective among evaluated. derived have potential serve as valuable references purposes land‐use planning catastrophe risk management.

Language: Английский

Citations

1