Sustainable groundwater management in coastal cities: Insights from groundwater potential and vulnerability using ensemble learning and knowledge-driven models

Journal of Cleaner Production, Год журнала: 2024, Номер 442, С. 141152 - 141152

Опубликована: Фев. 1, 2024

Язык: Английский

---

Enhancing the exploitation of natural resources for green energy: An application of LSTM-based meta-model for aluminum prices forecasting

Resources Policy, Год журнала: 2024, Номер 92, С. 105014 - 105014

Опубликована: Май 1, 2024

Язык: Английский

---

Fostering sustainable mining practices in rock blasting: Assessment of blast toe volume prediction using comparative analysis of hybrid ensemble machine learning techniques

Journal of Safety and Sustainability, Год журнала: 2024, Номер 1(2), С. 75 - 88

Опубликована: Май 24, 2024

Blast toe volume, pivotal in explosive engineering, underpins energy efficient utilization, blast safety and mine production sustainability. While current research explores the use of artificial intelligence (AI) model to minaimize volume production, gaps persist understanding application ensemble learning algorithm techniques like hybrid voting addressing problem. Bridging these promises enhanced optimization blasting operations. This study Performs AI enhance Toe prediction robustness by leveraging diverse algorithms, mitigating biases, optimizing accuracy. The combines separate models, looks for ways that approaches can work together, improves accuracy through group order give more complete information accurate predictions estimating different approaches. To develop 457 data was collected at Anguran lead zinc Iran. developed models assessed using nine indices compare their performance. understand input relationship, Multicollinearity, Spearman, Kendall correlation analyses show there is a strong link between size charge per delay. Findings from analysis showed Light Gradient Boosting Machine (LightGBM) most 8 traditional with R2 values 0.9004 training dataset 0.8625 testing dataset. Hybrid 6 model, which LightGBM Classification Regression Trees (CART) achieved highest scores 0.9473 phase 0.9467 phase. Voting consisting LightGBM, (GBM), Decision tree, Ensemble Random Forest, CatBoost, CART, AdaBoost, XGBoost, had greatest 0.9876 0.97265 both stages. Using novel modelling tools forecast this allows resource extraction optimization, decreases environmental disturbance smoothening, safety, supporting sustainable mining practices long-term

Язык: Английский

---

Exploring the spatial patterns of landslide susceptibility assessment using interpretable Shapley method: Mechanisms of landslide formation in the Sichuan-Tibet region

Journal of Environmental Management, Год журнала: 2024, Номер 366, С. 121921 - 121921

Опубликована: Июль 24, 2024

Язык: Английский

---

Optimization of SVR and CatBoost models using metaheuristic algorithms to assess landslide susceptibility

Scientific Reports, Год журнала: 2024, Номер 14(1)

Опубликована: Окт. 22, 2024

In this study, a landslide susceptibility assessment is performed by combining two machine learning regression algorithms (MLRA), such as support vector (SVR) and categorical boosting (CatBoost), with population-based optimization algorithms, grey wolf optimizer (GWO) particle swarm (PSO), to evaluate the potential of relatively new algorithm impact that can have on performance models. The Kerala state in India has been chosen test site due large number recorded incidents recent past. study started 18 predisposing factors, which were reduced 14 after multi-approach feature selection technique. Six models implemented compared using alone each them algorithms: SVR, CatBoost, SVR-PSO, CatBoost-PSO, SVR-GWO, CatBoost-GWO. resulting maps validated an independent dataset. rankings, based area under receiver operating characteristic curve (AUC) metric, are follows: CatBoost-GWO (AUC = 0.910) had highest performance, followed CatBoost-PSO 0.909), CatBoost 0.899), SVR-GWO 0.868), SVR-PSO 0.858), SVR 0.840). Other validation statistics corroborated these outcomes, Friedman Wilcoxon-signed rank tests verified statistical significance Our case showed outperformed both optimized or not; introduction significantly improves results models, GWO being slightly more effective than PSO. However, cannot drastically alter model, highlighting importance setting up rigorous model since early steps any research.

Язык: Английский

---

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

Journal of Rock Mechanics and Geotechnical Engineering, Год журнала: 2024, Номер 16(8), С. 3221 - 3232

Опубликована: Фев. 7, 2024

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

Язык: Английский

---

Integrating a multi-dimensional deep convolutional neural network with optimized sample selection for landslide susceptibility assessment

Geo-spatial Information Science, Год журнала: 2025, Номер unknown, С. 1 - 21

Опубликована: Янв. 17, 2025

To address the errors of negative samples in landslide susceptibility modeling and traditional methods exploring regularities hidden evaluation factors, this paper proposes a stacking one- three-dimensional Convolutional Neural Network (Stacking-1D-3D-CNN) assessment method considering sample optimization selection. First, order to select rationally, adopts Relative Frequency Ratio combined with Certainty Factor Method (RFR-CFM) determine samples; secondly, Stacking-1D-3D-CNN proposed is RFR-CFM for first time assessment. In work, determined by Information Quality Model (IQM) were historical disaster points form total sample, modeled at different ratios. Finally, it compared several other models terms hazard zoning results, prone zone statistics, model performance. The findings show that degree spatial aggregation training testing has much greater impact on accuracy than their proportions. Furthermore, models, RFR-CFM-Stacking-1D-3D-CNN highest AUC value, precision, recall, F-score, accuracy, which are 0.95, 0.83, 0.89, 0.85, 84.76%, respectively, lowest RMSE MAE, 0.39 0.15, respectively. This proves selection method's rationality model's effectiveness.

Язык: Английский

---

Spatiotemporal variation and driving factors of vegetation net primary productivity in the Guanzhong Plain Urban Agglomeration, China from 2001 to 2020

Journal of Arid Land, Год журнала: 2025, Номер 17(1), С. 74 - 92

Опубликована: Янв. 1, 2025

Язык: Английский

---

Landslide susceptibility mapping using artificial intelligence models: a case study in the Himalayas

Landslides, Год журнала: 2025, Номер unknown

Опубликована: Фев. 13, 2025

Язык: Английский

---

Machine learning for screw design in single‐screw extrusion

Polymer Engineering and Science, Год журнала: 2025, Номер unknown

Опубликована: Март 10, 2025

Abstract Artificial intelligence (AI) methods have significantly impacted various areas of technology, particularly in fields where large datasets are available. Screw designs proprietary, and there is very limited information available the open literature. In this study, we generated a dataset 232 using computer simulation software for screw extrusion, involving solids transport, melting, melt pumping. The parameters (features) outputs (targets) were introduced into four powerful machine learning (ML) algorithms. capabilities algorithms assessed by comparing predictions each to corresponding results simulations. Three demonstrated satisfactory performance, with best‐performing one being further tested on an “unseen” dataset, which involved 75 mm another 127 diameter. A machine‐learning technique called Permutation Feature Importance (PFI) was used identify features (parameters) greatest impact predictions. It suggested that same ML methodologies could be applied existing real designs. Highlights Dataset obtained from software. Four employed. Assessment based training testing data. Identification having impact. Satisfactory mass flow rate, exit temperature, melting length, more.

Язык: Английский

---