ResNet18 facial feature extraction algorithm improved based on hybrid domain attention mechanism DOI Creative Commons

Yuwen Mei

PLoS ONE, Journal Year: 2025, Volume and Issue: 20(3), P. e0319921 - e0319921

Published: March 19, 2025

In the research of face recognition technology, traditional methods usually show poor accuracy and insufficient generalization ability when faced with complex scenes such as lighting changes, posture changes skin color diversity. To solve these problems, based on improvement adaptive boosting to improve detection, study proposes a residual network 18-layer feature extraction algorithm hybrid domain attention mechanism algorithm. The introduces channel-domain spatial-domain enhance image features. outcomes indicated that proposed method multiple datasets, labeled field celebrity facial attribute datasets exceeded 98.34% reached up 99.64%, which was better than current state-of-the-art methods. After combining channel spatial mechanism, false detection rate low 2.50%, lower other addition enhancing recognition's robustness accuracy, work offers fresh concepts resources for potential uses in intricate scenarios future.

Language: Английский

Optimized hybrid ensemble learning approaches applied to very short-term load forecasting DOI Creative Commons
Marcos Yamasaki, Roberto Zanetti Freire, Laio Oriel Seman

et al.

International Journal of Electrical Power & Energy Systems, Journal Year: 2023, Volume and Issue: 155, P. 109579 - 109579

Published: Oct. 26, 2023

The significance of accurate short-term load forecasting (STLF) for modern power systems' efficient and secure operation is paramount. This task intricate due to cyclicity, non-stationarity, seasonality, nonlinear consumption time series data characteristics. rise accessibility in the industry has paved way machine learning (ML) models, which show potential enhance STLF accuracy. paper presents a novel hybrid ML model combining Gradient Boosting Regressor (GBR), Extreme (XGBoost), k-Nearest Neighbors (kNN), Support Vector Regression (SVR), examining both standalone integrated, coupled with signal decomposition techniques like STL, EMD, EEMD, CEEMDAN, EWT. Through Automated Machine Learning (AutoML), these models are integrated their hyperparameters optimized, predicting each component using from two sources: National Operator Electric System (ONS) Independent Operators New England (ISO-NE), boosting prediction capacity. For 2019 ONS dataset, EWT XGBoost yielded best results very (VSTLF) an RMSE 1,931.8 MW, MAE 1,564.9 MAPE 2.54%. These findings highlight necessity diverse approaches VSTLF problem, emphasizing adaptability strength combined techniques.

Language: Английский

Citations

85
Explainable machine learning model for liquefaction potential assessment of soils using XGBoost-SHAP DOI
Kaushik Jas, G. R. Dodagoudar

Soil Dynamics and Earthquake Engineering, Journal Year: 2022, Volume and Issue: 165, P. 107662 - 107662

Published: Nov. 30, 2022

Language: Английский

Citations

83
Predicting occurrence of liquefaction-induced lateral spreading using gradient boosting algorithms integrated with particle swarm optimization: PSO-XGBoost, PSO-LightGBM, and PSO-CatBoost DOI
Selçuk Demir, Emrehan Kutluğ Şahin

Acta Geotechnica, Journal Year: 2023, Volume and Issue: 18(6), P. 3403 - 3419

Published: Jan. 2, 2023

Language: Английский

Citations

68
Short-Term Rockburst Damage Assessment in Burst-Prone Mines: An Explainable XGBOOST Hybrid Model with SCSO Algorithm DOI

Yingui Qiu,

Jian Zhou

Rock Mechanics and Rock Engineering, Journal Year: 2023, Volume and Issue: 56(12), P. 8745 - 8770

Published: Sept. 2, 2023

Language: Английский

Citations

56
Predictive modeling of land surface temperature (LST) based on Landsat-8 satellite data and machine learning models for sustainable development DOI

Chaitanya B. Pande,

Johnbosco C. Egbueri, Romulus Costache

et al.

Journal of Cleaner Production, Journal Year: 2024, Volume and Issue: 444, P. 141035 - 141035

Published: Feb. 8, 2024

Language: Английский

Citations

35
Evaluation and analysis of liquefaction potential of gravelly soils using explainable probabilistic machine learning model DOI
Kaushik Jas, Sujith Mangalathu, G. R. Dodagoudar

et al.

Computers and Geotechnics, Journal Year: 2024, Volume and Issue: 167, P. 106051 - 106051

Published: Jan. 8, 2024

Language: Английский

Citations

20
Exploring forest fire susceptibility and management strategies in Western Himalaya: Integrating ensemble machine learning and explainable AI for accurate prediction and comprehensive analysis DOI Creative Commons
Hoang Thi Hang, Javed Mallick, Saeed Alqadhi

et al.

Environmental Technology & Innovation, Journal Year: 2024, Volume and Issue: 35, P. 103655 - 103655

Published: May 5, 2024

Forest fires pose a significant threat to ecosystems and socio-economic activities, necessitating the development of accurate predictive models for effective management mitigation. In this study, we present novel machine learning approach combined with Explainable Artificial Intelligence (XAI) techniques predict forest fire susceptibility in Nainital district. Our innovative methodology integrates several robust — AdaBoost, Gradient Boosting Machine (GBM), XGBoost Random Deep Neural Network (DNN) as meta-model stacking framework. This not only utilises individual strengths these models, but also improves overall prediction performance reliability. By using XAI techniques, particular SHAP (SHapley Additive exPlanations) LIME (Local Interpretable Model-agnostic Explanations), improve interpretability provide insights into decision-making processes. results show effectiveness ensemble model categorising different zones: very low, moderate, high high. particular, identified extensive areas susceptibility, precision, recall F1 values underpinning their effectiveness. These achieved ROC AUC above 0.90, performing exceptionally well an 0.94. The are remarkably inclusion confidence intervals most important metrics all emphasises robustness reliability supports practical use management. Through summary plots, analyze global variable importance, revealing annual rainfall Evapotranspiration (ET) key factors influencing susceptibility. Local analysis consistently highlights importance rainfall, ET, distance from roads across models. study fills research gap by providing comprehensive interpretable modelling that our ability effectively manage risk is consistent environmental protection sustainable goals.

Language: Английский

Citations

17
The Role of Utilizing Artificial Intelligence and Renewable Energy in Reaching Sustainable Development Goals DOI
Fatma M. Talaat, A.E. Kabeel,

Warda M. Shaban

et al.

Renewable Energy, Journal Year: 2024, Volume and Issue: 235, P. 121311 - 121311

Published: Sept. 7, 2024

Language: Английский

Citations

11
An Optimal House Price Prediction Algorithm: XGBoost DOI Creative Commons
Hemlata Sharma,

Hitesh Harsora,

Bayode Ogunleye

et al.

Analytics, Journal Year: 2024, Volume and Issue: 3(1), P. 30 - 45

Published: Jan. 2, 2024

An accurate prediction of house prices is a fundamental requirement for various sectors including real estate and mortgage lending. It widely recognized that property value not solely determined by its physical attributes but significantly influenced surrounding neighbourhood. Meeting the diverse housing needs individuals while balancing budget constraints primary concern developers. To this end, we addressed price problem as regression task thus employed machine learning techniques capable expressing significance independent variables. We made use dataset Ames City in Iowa, USA to compare support vector regressor, random forest XGBoost, multilayer perceptron multiple linear algorithms prediction. Afterwards, identified key factors influence costs. Our results show XGBoost best performing model

Language: Английский

Citations

9
Surface water quality prediction in the lower Thoubal river watershed, India: A hyper-tuned machine learning approach and DNN-based sensitivity analysis DOI
Md Hibjur Rahaman, Haroon Sajjad,

Shabina Hussain

et al.

Journal of environmental chemical engineering, Journal Year: 2024, Volume and Issue: 12(3), P. 112915 - 112915

Published: May 3, 2024

Language: Английский

Citations

9