Investigation of Landslide Susceptibility Decision Mechanisms in Different Ensemble-Based Machine Learning Models with Various Types of Factor Data DOI Open Access
Jiakai Lu, Chao Ren, Weiting Yue

et al.

Sustainability, Journal Year: 2023, Volume and Issue: 15(18), P. 13563 - 13563

Published: Sept. 11, 2023

Machine learning (ML)-based methods of landslide susceptibility assessment primarily focus on two dimensions: accuracy and complexity. The complexity is not only influenced by specific model frameworks but also the type modeling data. Therefore, considering impact factor data types model’s decision-making mechanism holds significant importance in assessing regional characteristics conducting risk warnings given achievement good predictive performance for using excellent ML methods. models coupled with different machine was explained this study utilizing Shapley Additive exPlanations (SHAP) method. Furthermore, a comparative analysis carried out to examine differential effects diverse identical factors predictions. area selected Cenxi, Guangxi, where geographic spatial database constructed combining 23 conditioning 214 samples from region. Initially, were standardized five conditional probability models, frequency ratio (FR), information value (IV), certainty (CF), evidential belief function (EBF), weights evidence (WOE), based arrangement landslides. This led formation six databases initial Subsequently, ensemble-based methods, random forest (RF) XGBoost, utilized build predicting susceptibility. Various evaluation metrics employed compare capabilities determined optimal model. Simultaneously, conducted interpretable SHAP method intrinsic mechanisms explaining comparing impacts prediction results. results illustrated that XGBoost-CF CF values exhibited best stability yielded more reasonable zoning, thus identified as global interpretation revealed slope most crucial influencing landslides, its interaction other collectively contributed occurrences. differences internal same manifested extent influence dependency factors, providing an explanation reasons behind higher Through comprehensive local analyzing sample characteristics, errors can be summarized, thereby reference framework constructing accurate rational facilitating warning management.

Language: Английский

Explainable artificial intelligence in disaster risk management: Achievements and prospective futures DOI Creative Commons
Saman Ghaffarian, Firouzeh Taghikhah, Holger R. Maier

et al.

International Journal of Disaster Risk Reduction, Journal Year: 2023, Volume and Issue: 98, P. 104123 - 104123

Published: Nov. 1, 2023

Disasters can have devastating impacts on communities and economies, underscoring the urgent need for effective strategic disaster risk management (DRM). Although Artificial Intelligence (AI) holds potential to enhance DRM through improved decision-making processes, its inherent complexity "black box" nature led a growing demand Explainable AI (XAI) techniques. These techniques facilitate interpretation understanding of decisions made by models, promoting transparency trust. However, current state XAI applications in DRM, their achievements, challenges they face remain underexplored. In this systematic literature review, we delve into burgeoning domain XAI-DRM, extracting 195 publications from Scopus ISI Web Knowledge databases, selecting 68 detailed analysis based predefined exclusion criteria. Our study addresses pertinent research questions, identifies various hazard types, components, methods, uncovers limitations these approaches, provides synthesized insights explainability effectiveness decision-making. Notably, observed significant increase use 2022 2023, emphasizing interpretability. Through rigorous methodology, offer key directions that serve as guide future studies. recommendations highlight importance multi-hazard analysis, integration early warning systems digital twins, incorporation causal inference methods strategy planning effectiveness. This serves beacon researchers practitioners alike, illuminating intricate interplay between revealing profound solutions revolutionizing management.

Language: Английский

Citations

57
A systematic review of trustworthy artificial intelligence applications in natural disasters DOI Creative Commons
A. S. Albahri, Yahya Layth Khaleel, Mustafa Abdulfattah Habeeb

et al.

Computers & Electrical Engineering, Journal Year: 2024, Volume and Issue: 118, P. 109409 - 109409

Published: June 29, 2024

Artificial intelligence (AI) holds significant promise for advancing natural disaster management through the use of predictive models that analyze extensive datasets, identify patterns, and forecast potential disasters. These facilitate proactive measures such as early warning systems (EWSs), evacuation planning, resource allocation, addressing substantial challenges associated with This study offers a comprehensive exploration trustworthy AI applications in disasters, encompassing management, risk assessment, prediction. research is underpinned by an review reputable sources, including Science Direct (SD), Scopus, IEEE Xplore (IEEE), Web (WoS). Three queries were formulated to retrieve 981 papers from earliest documented scientific production until February 2024. After meticulous screening, deduplication, application inclusion exclusion criteria, 108 studies included quantitative synthesis. provides specific taxonomy disasters explores motivations, challenges, recommendations, limitations recent advancements. It also overview techniques developments using explainable artificial (XAI), data fusion, mining, machine learning (ML), deep (DL), fuzzy logic, multicriteria decision-making (MCDM). systematic contribution addresses seven open issues critical solutions essential insights, laying groundwork various future works trustworthiness AI-based management. Despite benefits, persist In these contexts, this identifies several unused used areas disaster-based theory, collects ML, DL techniques, valuable XAI approach unravel complex relationships dynamics involved utilization fusion processes related Finally, extensively analyzed ethical considerations, bias, consequences AI.

Language: Английский

Citations

42
Landslide Susceptibility Mapping Based on Interpretable Machine Learning from the Perspective of Geomorphological Differentiation DOI Creative Commons
Deliang Sun,

D C Chen,

Jialan Zhang

et al.

Land, Journal Year: 2023, Volume and Issue: 12(5), P. 1018 - 1018

Published: May 5, 2023

(1) Background: The aim of this paper was to study landslide susceptibility mapping based on interpretable machine learning from the perspective topography differentiation. (2) Methods: This selects three counties (Chengkou, Wushan and Wuxi counties) in northeastern Chongqing, delineated as corrosion layered high middle mountain region (Zone I), (Wulong, Pengshui Shizhu southeastern mountainous strong karst gorges II), area. used a Bayesian optimization algorithm optimize parameters LightGBM XGBoost models construct evaluation for each two regions. model with accuracy selected according indicators order establish mapping. SHAP then explore formation mechanisms different landforms both global local perspective. (3) Results: AUC values test set mode Zones I II are 0.8525 0.8859, respectively, those 0.8214 0.8375, respectively. shows that has prediction regard landforms. Under landform types, elevation, land use, incision depth, distance road average annual rainfall were common dominant factors contributing most decision making at sites; fault river have degrees influence under types. (4) Conclusions: optimized LightGBM-SHAP is suitable analysis types landscapes, namely region, gorges, can be internal decision-making mechanism levels, which makes results more realistic transparent. beneficial selection index system early prevention control hazards, provide reference potential hazard-prone areas research.

Language: Английский

Citations

34
Explainable machine learning for the prediction and assessment of complex drought impacts DOI
Beichen Zhang, Fatima K. Abu Salem, Michael J. Hayes

et al.

The Science of The Total Environment, Journal Year: 2023, Volume and Issue: 898, P. 165509 - 165509

Published: July 17, 2023

Language: Английский

Citations

29
Performance comparison of landslide susceptibility mapping under multiple machine-learning based models considering InSAR deformation: a case study of the upper Jinsha River DOI Creative Commons
Jiaming Yao, Xin Yao, Zheng Zhao

et al.

Geomatics Natural Hazards and Risk, Journal Year: 2023, Volume and Issue: 14(1)

Published: June 6, 2023

Landslide susceptibility mapping (LSM) comprehensively evaluates the spatial probability of landslide occurrence by using different environmental factors. However, most evaluation methods ignore dynamic characteristic factors landslides, which makes it difficult to obtain reliable prediction results. Taking upper reaches Jinsha River as study area, this article introduces deformation data into model and proposes an improved method. Four kinds machine learning models were constructed collecting 20 related The accuracy is compared, performance improvement information are evaluated. results show that Random Forest XGBoost better than SVM logistic regression model. obviously after InSAR introduced. 96.9 93.19% areas reasonably classified high or very risk levels. Compared with calculation result traditional model, proportion pixels in area increased 2.97 1.13%, respectively. In addition, percentage from 15.45 16.23% 18.73 21.89%, 0.793 0.878 0.776 0.812, respectively, AUC 0.9 1.7%, SHAP feature importance analysis reveals rainfall, aspect, temperature NDVI main influencing River.

Language: Английский

Citations

24
Rapid Landslide Extraction from High-Resolution Remote Sensing Images Using SHAP-OPT-XGBoost DOI Creative Commons

Na Lin,

Di Zhang, Shanshan Feng

et al.

Remote Sensing, Journal Year: 2023, Volume and Issue: 15(15), P. 3901 - 3901

Published: Aug. 7, 2023

Landslides, the second largest geological hazard after earthquakes, result in significant loss of life and property. Extracting landslide information quickly accurately is basis disaster prevention. Fengjie County, Chongqing, China, a typical landslide-prone area Three Gorges Reservoir Area. In this study, we newly integrate Shapley Additive Explanation (SHAP) Optuna (OPT) hyperparameter tuning into four basic machine learning algorithms: Gradient Boosting Decision Tree (GBDT), Extreme (XGBoost), Light Machine (LightGBM), (AdaBoost). We construct new models (SHAP-OPT-GBDT, SHAP-OPT-XGBoost, SHAP-OPT-LightGBM, SHAP-OPT-AdaBoost) apply to extraction for first time. Firstly, high-resolution remote sensing images were preprocessed, non-landslide samples constructed, an initial feature set with 48 features was built. Secondly, SHAP used select contributions, important selected. Finally, Optuna, Bayesian optimization technique, utilized automatically models’ best hyperparameters. The experimental results show that accuracy (ACC) these SHAP-OPT above 92% training time less than 1.3 s using mediocre computational hardware. Furthermore, SHAP-OPT-XGBoost achieved highest (96.26%). Landslide distribution County from 2013 2020 can be extracted by quickly.

Language: Английский

Citations

24
River ecological flow early warning forecasting using baseflow separation and machine learning in the Jiaojiang River Basin, Southeast China DOI
Hao Chen, Saihua Huang, Yue‐Ping Xu

et al.

The Science of The Total Environment, Journal Year: 2023, Volume and Issue: 882, P. 163571 - 163571

Published: April 21, 2023

Language: Английский

Citations

21
A hybrid MCDA approach for delineating sites suitable for artificial groundwater recharge using drywells DOI
Rachid Mohamed Mouhoumed, Ömer Ekmekcioğlu, Mehmet Özger

et al.

Journal of Hydrology, Journal Year: 2023, Volume and Issue: 620, P. 129387 - 129387

Published: March 11, 2023

Language: Английский

Citations

18
Refined Landslide Susceptibility Mapping by Integrating the SHAP-CatBoost Model and InSAR Observations: A Case Study of Lishui, Southern China DOI Creative Commons

Zhaowei Yao,

Meihong Chen, Jiewei Zhan

et al.

Applied Sciences, Journal Year: 2023, Volume and Issue: 13(23), P. 12817 - 12817

Published: Nov. 29, 2023

Landslide susceptibility mapping based on static influence factors often exhibits issues of low accuracy and classification errors. To enhance the mapping, this study proposes a refined approach that integrates categorical boosting (CatBoost) with small baseline subset interferometric synthetic-aperture radar (SBAS-InSAR) results, achieving more precise detailed mapping. We utilized optical remote sensing images, information value (IV) model, fourteen influencing (elevation, slope, aspect, roughness, profile curvature, plane lithology, distance to faults, land use type, normalized difference vegetation index (NDVI), topographic wetness (TWI), rivers, roads, annual precipitation) establish IV-CatBoost landslide method. Subsequently, Sentinel-1A ascending data from January 2021 March 2023 were derive deformation rates within city Lishui in southern region China. Based outcomes derived SBAS-InSAR, discernment matrix was formulated rectify inaccuracies partitioned regions, leading creation CatBoost integration (IVCI) model. In end, we interpretations alongside surface deformations obtained SBAS-InSAR cross-verify excellence IVCI. Research findings indicate distinct enhancement levels across 165,784 grids (149.20 km2) following correction. The enhanced classes spectral characteristics images closely correspond trends cumulative deformation, reflecting high level consistency field-based conditions. These improved classifications effectively refinement proposed paper enhances prediction accuracy, providing valuable technical reference for hazard prevention control region.

Language: Английский

Citations

18
Multi-hazard could exacerbate in coastal Bangladesh in the context of climate change DOI
Mahfuzur Rahman, Shufeng Tian,

Md Sakib Hasan Tumon

et al.

Journal of Cleaner Production, Journal Year: 2024, Volume and Issue: 457, P. 142289 - 142289

Published: April 23, 2024

Language: Английский

Citations

7