Intrusion detection in Internet of Things(IoT)environments presents challenges due to heterogeneous devices,diverse attack vectors,and highly imbalanced datasets.Existing research on the ToN-IoT dataset has largely em...Intrusion detection in Internet of Things(IoT)environments presents challenges due to heterogeneous devices,diverse attack vectors,and highly imbalanced datasets.Existing research on the ToN-IoT dataset has largely emphasized binary classification and single-model pipelines,which often showstrong performance but limited generalizability,probabilistic reliability,and operational interpretability.This study proposes a stacked ensemble deep learning framework that integrates random forest,extreme gradient boosting,and a deep neural network as base learners,with CatBoost as the meta-learner.On the ToN-IoT Linux process dataset,the model achieved near-perfect discrimination(macro area under the curve=0.998),robust calibration,and superior F1-scores compared with standalone classifiers.Interpretability was achieved through SHapley Additive exPlanations–based feature attribution,which highlights actionable drivers ofmalicious behavior,such as command-line patterns,process scheduling anomalies,and CPU usage spikes,and aligns these indicators with MITRE ATT&CK tactics and techniques.Complementary analyses,including cumulative lift and sensitivity-specificity trade-offs,revealed the framework’s suitability for deployment in security operations centers,where calibrated risk scores,transparent explanations,and resource-aware triage are essential.These contributions bridge methodological rigor in artificial intelligence/machine learning with operational priorities in cybersecurity,delivering a scalable and explainable intrusion detection system suitable for real-world deployment in IoT environments.展开更多
Predicting concrete compressive strength with limited data remains a critical challenge in civil engineering.This study proposes a novel framework integrating Model-Agnostic Meta-Learning(MAML)with SHAP(Shapley Additi...Predicting concrete compressive strength with limited data remains a critical challenge in civil engineering.This study proposes a novel framework integrating Model-Agnostic Meta-Learning(MAML)with SHAP(Shapley Additive Explanations)to improve predictive accuracy and interpretability in data-scarce scenarios.Unlike conventional machine learning models that require extensive data,the MAML-based approach enables rapid adaptation to new tasks using minimal samples,offering robust generalization in few-shot learning contexts.The proposed pipeline includes structured preprocessing,normalization,a neural network-based meta-learning core,and SHAP-based feature attribution.A curated dataset of 430 samples was used,focusing on 28-day compressive strength,with input features including cement,water,aggregates,admixtures,and age.Compared to standard models like XGBoost and Random Forest,the MAML framework achieved superior performance,with MAE=3.56 MPa,RMSE=5.55 MPa,and R^(2)=0.913.SHAP analysis revealed nonlinear interactions and dominant factors like water-cement ratio,curing age,and aggregate content.Statistical validation via the Wilcoxon Signed-Rank Test confirmed the significance of the model’s improvements(p<0.05).Furthermore,SHAP insights closely align with domain knowledge and mix design principles,enhancing model transparency for practical application.This work demonstrates the applicability of meta-learning in civil engineering and provides a scalable,interpretable solution for strength prediction in real-world,data-limited conditions.展开更多
Acquiring accurate molecular-level information about petroleum is crucial for refining and chemical enterprises to implement the“selection of the optimal processing route”strategy.With the development of data predic...Acquiring accurate molecular-level information about petroleum is crucial for refining and chemical enterprises to implement the“selection of the optimal processing route”strategy.With the development of data prediction systems represented by machine learning,it has become possible for real-time prediction systems of petroleum fraction molecular information to replace analyses such as gas chromatography and mass spectrometry.However,the biggest difficulty lies in acquiring the data required for training the neural network.To address these issues,this work proposes an innovative method that utilizes the Aspen HYSYS and full two-dimensional gas chromatography-time-of-flight mass spectrometry to establish a comprehensive training database.Subsequently,a deep neural network prediction model is developed for heavy distillate oil to predict its composition in terms of molecular structure.After training,the model accurately predicts the molecular composition of catalytically cracked raw oil in a refinery.The validation and test sets exhibit R2 values of 0.99769 and 0.99807,respectively,and the average relative error of molecular composition prediction for raw materials of the catalytic cracking unit is less than 7%.Finally,the SHAP(SHapley Additive ExPlanation)interpretation method is used to disclose the relationship among different variables by performing global and local weight comparisons and correlation analyses.展开更多
This paper proposes a machine learning framework for accurately predicting the aerodynamic lift-to-drag ratio(CL/CD)of multi-stepped airfoils under varied flow conditions.Experimental wind-tunnel data were collected f...This paper proposes a machine learning framework for accurately predicting the aerodynamic lift-to-drag ratio(CL/CD)of multi-stepped airfoils under varied flow conditions.Experimental wind-tunnel data were collected for multiple step configurations,and a stacked ensemble model combining XGBoost,Support Vector Regression(SVR),and K-Nearest Neighbors(KNN)with a Random Forest meta-learner was developed for prediction.The proposed model achieved a test R^(2)of 0.9951 and a tenfold cross-validation R^(2) of 0.9872±0.0043,demonstrating superior accuracy compared to individual regressors.This approach provides a fast,data-driven alternative to conventional CFD simulations,enabling reliable prediction of aerodynamic performance and efficient airfoil optimization.展开更多
Urban Nitrogen Oxide(NO_(x))air pollution poses significant public health risks,and therefore spatiotemporal knowledge of NO_(x) is crucial for air quality regulation.However,few studies have examined long-term NO_(x)...Urban Nitrogen Oxide(NO_(x))air pollution poses significant public health risks,and therefore spatiotemporal knowledge of NO_(x) is crucial for air quality regulation.However,few studies have examined long-term NO_(x) dynamics in Manchester by integrating spatial,temporal,and mechanistic perspectives.This study investigates the long-term trends and drivers of NO and NO_(2) in the urban atmosphere of Manchester from 2015 to 2025.Data from five AURN monitoring sites,ERA5 meteorological reanalysis datasets,and UK Department for Transport traffic statistics were analysed using linear regression for trend estimation,seasonal decomposition,and spatial pattern analysis.A hybrid statistical-machine learning framework was additionally employed,combining Ordinary Least Squares(OLS)regression with XGBoost models interpreted through Shapley Additive Explanations(SHAP).The results indicate statistically significant declining trends in both pollutants,with average annual decreases of approximately 4.8%,and dramatic short-term reductions during COVID-19 lockdowns,highlighting the dominant influence of traffic.Seasonal patterns persisted,with winter concentrations 1.9 times greater than summer levels,and spatial analysis revealed strong NO_(2) heterogeneity among monitoring sites.Machine learning models performed substantially better than linear regression(R^(2)=0.475 vs.0.29),and SHAP analysis showed ozone,boundary layer height,and temperature as the main drivers of NO_(2) variations.Overall,the findings confirm substantial air-quality improvements while revealing nonlinear processes in urban pollution dynamics,supporting continued emission-reduction policies and enhanced monitoring strategies.展开更多
Accurate real-time electricity price forecasting is of critical importance for market participants seeking to optimize energy procurement,dispatch scheduling,and arbitrage strategies in liberalized electricity markets...Accurate real-time electricity price forecasting is of critical importance for market participants seeking to optimize energy procurement,dispatch scheduling,and arbitrage strategies in liberalized electricity markets.However,existing forecasting approaches suffer from several key limitations:(1)conventional statistical models fail to capture the complex nonlinear interactions among generation mix,load demand,and temporal variables that collectively drive price dynamics;(2)single-model approaches lack robustness and are sensitive to overfitting,limiting their generalizability across diverse market conditions;(3)the interpretability of black-box prediction models remains insufficient,hindering the practical deployment of data-driven forecasting systems in operational decision-making.To address these challenges,this study proposes a comprehensive machine learning framework based on six tree-based ensemble models for hourly electricity price prediction in the Spanish electricity market.The proposed framework introduces three key contributions:(1)a systematic feature engineering pipeline incorporating lagged price variables,rolling statistics,and calendar-based temporal encodings;(2)a rigorous comparative evaluation of Decision Tree,Random Forest,Extra Trees,Gradient Boosting,XGBoost,and LightGBM under identical experimental conditions;(3)a SHAP-based interpretability analysis that quantifies feature contributions and interaction effects at both global and local levels.Experimental results on the ENTSO-E Spanish market dataset demonstrate that XGBoost achieves the best overall predictive performance,with an R²of 0.9660 and MAE of 1.5631€/MWh.展开更多
文摘Intrusion detection in Internet of Things(IoT)environments presents challenges due to heterogeneous devices,diverse attack vectors,and highly imbalanced datasets.Existing research on the ToN-IoT dataset has largely emphasized binary classification and single-model pipelines,which often showstrong performance but limited generalizability,probabilistic reliability,and operational interpretability.This study proposes a stacked ensemble deep learning framework that integrates random forest,extreme gradient boosting,and a deep neural network as base learners,with CatBoost as the meta-learner.On the ToN-IoT Linux process dataset,the model achieved near-perfect discrimination(macro area under the curve=0.998),robust calibration,and superior F1-scores compared with standalone classifiers.Interpretability was achieved through SHapley Additive exPlanations–based feature attribution,which highlights actionable drivers ofmalicious behavior,such as command-line patterns,process scheduling anomalies,and CPU usage spikes,and aligns these indicators with MITRE ATT&CK tactics and techniques.Complementary analyses,including cumulative lift and sensitivity-specificity trade-offs,revealed the framework’s suitability for deployment in security operations centers,where calibrated risk scores,transparent explanations,and resource-aware triage are essential.These contributions bridge methodological rigor in artificial intelligence/machine learning with operational priorities in cybersecurity,delivering a scalable and explainable intrusion detection system suitable for real-world deployment in IoT environments.
文摘Predicting concrete compressive strength with limited data remains a critical challenge in civil engineering.This study proposes a novel framework integrating Model-Agnostic Meta-Learning(MAML)with SHAP(Shapley Additive Explanations)to improve predictive accuracy and interpretability in data-scarce scenarios.Unlike conventional machine learning models that require extensive data,the MAML-based approach enables rapid adaptation to new tasks using minimal samples,offering robust generalization in few-shot learning contexts.The proposed pipeline includes structured preprocessing,normalization,a neural network-based meta-learning core,and SHAP-based feature attribution.A curated dataset of 430 samples was used,focusing on 28-day compressive strength,with input features including cement,water,aggregates,admixtures,and age.Compared to standard models like XGBoost and Random Forest,the MAML framework achieved superior performance,with MAE=3.56 MPa,RMSE=5.55 MPa,and R^(2)=0.913.SHAP analysis revealed nonlinear interactions and dominant factors like water-cement ratio,curing age,and aggregate content.Statistical validation via the Wilcoxon Signed-Rank Test confirmed the significance of the model’s improvements(p<0.05).Furthermore,SHAP insights closely align with domain knowledge and mix design principles,enhancing model transparency for practical application.This work demonstrates the applicability of meta-learning in civil engineering and provides a scalable,interpretable solution for strength prediction in real-world,data-limited conditions.
基金the National Natural Science Foundation of China(22108307)the Natural Science Foundation of Shandong Province(ZR2020KB006)the Outstanding Youth Fund of Shandong Provincial Natural Science Foundation(ZR2020YQ17).
文摘Acquiring accurate molecular-level information about petroleum is crucial for refining and chemical enterprises to implement the“selection of the optimal processing route”strategy.With the development of data prediction systems represented by machine learning,it has become possible for real-time prediction systems of petroleum fraction molecular information to replace analyses such as gas chromatography and mass spectrometry.However,the biggest difficulty lies in acquiring the data required for training the neural network.To address these issues,this work proposes an innovative method that utilizes the Aspen HYSYS and full two-dimensional gas chromatography-time-of-flight mass spectrometry to establish a comprehensive training database.Subsequently,a deep neural network prediction model is developed for heavy distillate oil to predict its composition in terms of molecular structure.After training,the model accurately predicts the molecular composition of catalytically cracked raw oil in a refinery.The validation and test sets exhibit R2 values of 0.99769 and 0.99807,respectively,and the average relative error of molecular composition prediction for raw materials of the catalytic cracking unit is less than 7%.Finally,the SHAP(SHapley Additive ExPlanation)interpretation method is used to disclose the relationship among different variables by performing global and local weight comparisons and correlation analyses.
基金provided by The Science,Technology&Innovation Funding Authority(STDF)in cooperation with The Egyptian Knowledge Bank(EKB).
文摘This paper proposes a machine learning framework for accurately predicting the aerodynamic lift-to-drag ratio(CL/CD)of multi-stepped airfoils under varied flow conditions.Experimental wind-tunnel data were collected for multiple step configurations,and a stacked ensemble model combining XGBoost,Support Vector Regression(SVR),and K-Nearest Neighbors(KNN)with a Random Forest meta-learner was developed for prediction.The proposed model achieved a test R^(2)of 0.9951 and a tenfold cross-validation R^(2) of 0.9872±0.0043,demonstrating superior accuracy compared to individual regressors.This approach provides a fast,data-driven alternative to conventional CFD simulations,enabling reliable prediction of aerodynamic performance and efficient airfoil optimization.
文摘Urban Nitrogen Oxide(NO_(x))air pollution poses significant public health risks,and therefore spatiotemporal knowledge of NO_(x) is crucial for air quality regulation.However,few studies have examined long-term NO_(x) dynamics in Manchester by integrating spatial,temporal,and mechanistic perspectives.This study investigates the long-term trends and drivers of NO and NO_(2) in the urban atmosphere of Manchester from 2015 to 2025.Data from five AURN monitoring sites,ERA5 meteorological reanalysis datasets,and UK Department for Transport traffic statistics were analysed using linear regression for trend estimation,seasonal decomposition,and spatial pattern analysis.A hybrid statistical-machine learning framework was additionally employed,combining Ordinary Least Squares(OLS)regression with XGBoost models interpreted through Shapley Additive Explanations(SHAP).The results indicate statistically significant declining trends in both pollutants,with average annual decreases of approximately 4.8%,and dramatic short-term reductions during COVID-19 lockdowns,highlighting the dominant influence of traffic.Seasonal patterns persisted,with winter concentrations 1.9 times greater than summer levels,and spatial analysis revealed strong NO_(2) heterogeneity among monitoring sites.Machine learning models performed substantially better than linear regression(R^(2)=0.475 vs.0.29),and SHAP analysis showed ozone,boundary layer height,and temperature as the main drivers of NO_(2) variations.Overall,the findings confirm substantial air-quality improvements while revealing nonlinear processes in urban pollution dynamics,supporting continued emission-reduction policies and enhanced monitoring strategies.
文摘Accurate real-time electricity price forecasting is of critical importance for market participants seeking to optimize energy procurement,dispatch scheduling,and arbitrage strategies in liberalized electricity markets.However,existing forecasting approaches suffer from several key limitations:(1)conventional statistical models fail to capture the complex nonlinear interactions among generation mix,load demand,and temporal variables that collectively drive price dynamics;(2)single-model approaches lack robustness and are sensitive to overfitting,limiting their generalizability across diverse market conditions;(3)the interpretability of black-box prediction models remains insufficient,hindering the practical deployment of data-driven forecasting systems in operational decision-making.To address these challenges,this study proposes a comprehensive machine learning framework based on six tree-based ensemble models for hourly electricity price prediction in the Spanish electricity market.The proposed framework introduces three key contributions:(1)a systematic feature engineering pipeline incorporating lagged price variables,rolling statistics,and calendar-based temporal encodings;(2)a rigorous comparative evaluation of Decision Tree,Random Forest,Extra Trees,Gradient Boosting,XGBoost,and LightGBM under identical experimental conditions;(3)a SHAP-based interpretability analysis that quantifies feature contributions and interaction effects at both global and local levels.Experimental results on the ENTSO-E Spanish market dataset demonstrate that XGBoost achieves the best overall predictive performance,with an R²of 0.9660 and MAE of 1.5631€/MWh.
