Explainable artificial intelligence in disaster risk management: Achievements and prospective futures

International Journal of Disaster Risk Reduction, Год журнала: 2023, Номер 98, С. 104123 - 104123

Опубликована: Ноя. 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.

Язык: Английский

---

Enhancing flood risk assessment through integration of ensemble learning approaches and physical-based hydrological modeling

Geomatics Natural Hazards and Risk, Год журнала: 2023, Номер 14(1)

Опубликована: Май 4, 2023

This study aims to examine three machine learning (ML) techniques, namely random forest (RF), LightGBM, and CatBoost for flooding susceptibility maps (FSMs) in the Vietnamese Vu Gia-Thu Bon (VGTB). The results of ML are compared with those rainfall-runoff model, different training dataset sizes utilized performance assessment. Ten independent factors assessed. An inventory map approximately 850 sites is based on several post-flood surveys. randomly split between (70%) testing (30%). AUC-ROC 97.9%, 99.5%, 99.5% CatBoost, RF, respectively. FSMs developed by methods show good agreement terms an extension flood inundation using model. models' showed 10–13% total area be highly susceptible flooding, consistent RRI's map. that downstream areas (both urbanized agricultural) under high very levels susceptibility. Additionally, input datasets tested determine least number data points having acceptable reliability. demonstrate can realistically predict FSMs, regardless samples.

Язык: Английский

---

Research on Water Resource Modeling Based on Machine Learning Technologies

Water, Год журнала: 2024, Номер 16(3), С. 472 - 472

Опубликована: Янв. 31, 2024

Water resource modeling is an important means of studying the distribution, change, utilization, and management water resources. By establishing various models, resources can be quantitatively described predicted, providing a scientific basis for management, protection, planning. Traditional hydrological observation methods, often reliant on experience statistical are time-consuming labor-intensive, frequently resulting in predictions limited accuracy. However, machine learning technologies enhance efficiency sustainability by analyzing extensive hydrogeological data, thereby improving optimizing utilization allocation. This review investigates application predicting aspects, including precipitation, flood, runoff, soil moisture, evapotranspiration, groundwater level, quality. It provides detailed summary algorithms, examines their technical strengths weaknesses, discusses potential applications modeling. Finally, this paper anticipates future development trends to

Язык: Английский

---

Explainability in wind farm planning: A machine learning framework for automatic site selection of wind farms

Energy Conversion and Management, Год журнала: 2024, Номер 309, С. 118441 - 118441

Опубликована: Апрель 25, 2024

Язык: Английский

---

A deep-learning-based tree species classification for natural secondary forests using unmanned aerial vehicle hyperspectral images and LiDAR

Ecological Indicators, Год журнала: 2024, Номер 159, С. 111608 - 111608

Опубликована: Янв. 28, 2024

Accurate tree species classification is essential for forest resource management and biodiversity assessment. However, classifying becomes challenging in natural secondary forests due to the difficulties outlining crown boundary. In this study, an object-based framework Experimental Forestry Farm of Northeast University, located Heilongjiang Province, China, was developed based on unmanned aerial vehicle (UAV) hyperspectral images (HSIs) UAV light detection ranging (LiDAR) data using convolutional neural networks (CNNs). The study area characterized by representative that encompass diverse species, such as Korean pine (Pinus koraiensis Sieb. et Zucc.), White birch (Betula platyphylla Suk.), Siberian elm (Ulmus pumila L.), Manchurian ash (Fraxinus mandshurica Rupr.). This included two key processes: (1) u-shaped network (U-net) algorithm employed with simple linear iterative clustering (SLIC) algorithm, is, U-SLIC individual delineation (ITCD), (2) performances one-dimensional CNN (1D-CNN), two-dimensional (2D-CNN), three-dimensional (3D-CNN) models were compared while investigating role attention mechanism (convolutional block module, CBAM) added (1D-/2D-/3D-CNN + CBAM). results showed obtained a satisfactory accuracy ITCD procedure, recall 0.92, precision 0.79, F-score 0.85. feature selection effectively enhanced models' classification. Furthermore, adding CBAM resulted overall (OA) improvements 0.08, 0.11, 0.09 1D-CNN, 2D-CNN, 3D-CNN, respectively. 1D-CNN model performed best OA 0.83 when utilizing selected HSI LiDAR features. highlighted utilization integration multiple deep-learning algorithms complex forests, serving prerequisites decisions, conservation, carbon stock estimation.

Язык: Английский

---

Revealing the nature of soil liquefaction using machine learning

Earth Science Informatics, Год журнала: 2025, Номер 18(2)

Опубликована: Янв. 21, 2025

Язык: Английский

---

Snow avalanche susceptibility mapping using novel tree-based machine learning algorithms (XGBoost, NGBoost, and LightGBM) with eXplainable Artificial Intelligence (XAI) approach

Stochastic Environmental Research and Risk Assessment, Год журнала: 2023, Номер 37(6), С. 2243 - 2270

Опубликована: Март 13, 2023

Язык: Английский

---

Uncertainty Reduction in Flood Susceptibility Mapping Using Random Forest and eXtreme Gradient Boosting Algorithms in Two Tropical Desert Cities, Shibam and Marib, Yemen

Remote Sensing, Год журнала: 2024, Номер 16(2), С. 336 - 336

Опубликована: Янв. 15, 2024

Flooding is a natural disaster that coexists with human beings and causes severe loss of life property worldwide. Although numerous studies for flood susceptibility modelling have been introduced, notable gap has the overlooked or reduced consideration uncertainty in accuracy produced maps. Challenges such as limited data, due to confidence bounds, overfitting problem are critical areas improving accurate models. We focus on mapping, mainly when there significant variation predictive relevance predictor factors. It also noted receiver operating characteristic (ROC) curve may not accurately depict sensitivity resulting map overfitting. Therefore, reducing was targeted increase improve processing time prediction. This study created spatial repository test models, containing data from historical flooding twelve topographic geo-environmental conditioning variables. Then, we applied random forest (RF) extreme gradient boosting (XGB) algorithms susceptibility, incorporating variable drop-off empirical loop function. The results showed function crucial method resolve model associated factors methods. approximately 8.42% 9.89% Marib City 9.93% 15.69% Shibam were highly vulnerable floods. Furthermore, this significantly contributes worldwide endeavors focused hazards linked disasters. approaches used can offer valuable insights strategies risks, particularly Yemen.

Язык: Английский

---

Urban waterlogging susceptibility assessment based on hybrid ensemble machine learning models: A case study in the metropolitan area in Beijing, China

Journal of Hydrology, Год журнала: 2024, Номер 630, С. 130695 - 130695

Опубликована: Янв. 23, 2024

Язык: Английский

---

Artificial Intelligence in Agricultural Mapping: A Review

Agriculture, Год журнала: 2024, Номер 14(7), С. 1071 - 1071

Опубликована: Июль 3, 2024

Artificial intelligence (AI) plays an essential role in agricultural mapping. It reduces costs and time increases efficiency management activities, which improves the food industry. Agricultural mapping is necessary for resource requires technologies farming challenges. The AI applications gives its subsequent use decision-making. This study analyses AI’s current state through bibliometric indicators a literature review to identify methods, resources, geomatic tools, types, their management. methodology begins with bibliographic search Scopus Web of Science (WoS). Subsequently, data analysis establish scientific contribution, collaboration, trends. United States (USA), Spain, Italy are countries that produce collaborate more this area knowledge. Of studies, 76% machine learning (ML) 24% deep (DL) applications. Prevailing algorithms such as Random Forest (RF), Neural Networks (ANNs), Support Vector Machines (SVMs) correlate activities In addition, contributes associated production, disease detection, crop classification, rural planning, forest dynamics, irrigation system improvements.

Язык: Английский

---