Alzheimer’s disease(AD)is a significant challenge in modern healthcare,with early detection and accurate staging remaining critical priorities for effective intervention.While Deep Learning(DL)approaches have shown p...Alzheimer’s disease(AD)is a significant challenge in modern healthcare,with early detection and accurate staging remaining critical priorities for effective intervention.While Deep Learning(DL)approaches have shown promise in AD diagnosis,existing methods often struggle with the issues of precision,interpretability,and class imbalance.This study presents a novel framework that integrates DL with several eXplainable Artificial Intelligence(XAI)techniques,in particular attention mechanisms,Gradient-Weighted Class Activation Mapping(Grad-CAM),and Local Interpretable Model-Agnostic Explanations(LIME),to improve bothmodel interpretability and feature selection.The study evaluates four different DL architectures(ResMLP,VGG16,Xception,and Convolutional Neural Network(CNN)with attention mechanism)on a balanced dataset of 3714 MRI brain scans from patients aged 70 and older.The proposed CNN with attention model achieved superior performance,demonstrating 99.18%accuracy on the primary dataset and 96.64% accuracy on the ADNI dataset,significantly advancing the state-of-the-art in AD classification.The ability of the framework to provide comprehensive,interpretable results through multiple visualization techniques while maintaining high classification accuracy represents a significant advancement in the computational diagnosis of AD,potentially enabling more accurate and earlier intervention in clinical settings.展开更多
Diabetes imposes a substantial burden on global healthcare systems.Worldwide,nearly half of individuals with diabetes remain undiagnosed,while conventional diagnostic techniques are often invasive,painful,and expensiv...Diabetes imposes a substantial burden on global healthcare systems.Worldwide,nearly half of individuals with diabetes remain undiagnosed,while conventional diagnostic techniques are often invasive,painful,and expensive.In this study,we propose a noninvasive approach for diabetes detection using photoplethysmography(PPG),which is widely integrated into modern wearable devices.First,we derived velocity plethysmography(VPG)and acceleration plethysmography(APG)signals from PPG to construct multi-channel waveform representations.Second,we introduced a novel multiset spatiotemporal feature fusion framework that integrates hand-crafted temporal,statistical,and nonlinear features with recursive feature elimination and deep feature extraction using a one-dimensional statistical convolutional neural network(1DSCNN).Finally,we developed an interpretable diabetes detection method based on XGBoost,with explainable artificial intelligence(XAI)techniques.Specifically,SHapley Additive exPlanations(SHAP)and Local InterpretableModel-agnostic Explanations(LIME)were employed to identify and interpret potential digital biomarkers associated with diabetes.To validate the proposed method,we extended the publicly available Guilin People’s Hospital dataset by incorporating in-house clinical data from ten subjects,thereby enhancing data diversity.A subject-independent cross-validation strategy was applied to ensure that the testing subjects remained independent of the training data for robust generalization.Compared with existing state-of-the-art methods,our approach achieved superior performance,with an area under the curve(AUC)of 80.5±15.9%,sensitivity of 77.2±7.5%,and specificity of 64.3±18.2%.These results demonstrate that the proposed approach provides a noninvasive,interpretable,and accessible solution for diabetes detection using PPG signals.展开更多
The inherent black-box nature of machine learning(ML) models limits their interpretability and broader application in heavy precipitation forecasting. Evaluating the reliability of these models involves analyzing the ...The inherent black-box nature of machine learning(ML) models limits their interpretability and broader application in heavy precipitation forecasting. Evaluating the reliability of these models involves analyzing the link between predictions and predictors. In this study, ERA5 reanalysis data, CERES satellite observations, and ground-based meteorological observatories were utilized to compile more comprehensive multi-type predictors for developing a Bayesian optimized XGBoost model for the nowcasting of heavy precipitation in the Guangdong-Hong Kong-Macao Greater Bay Area during the pre-summer rainy season. A comparison of model performance with different combinations of input features and classical machine learning algorithms demonstrated that the Bayesian optimized XGBoost model achieved the best overall performance, with an average Critical Success Index of 68.30%. Permutation Importance(PI) and shapley Additive Explanations(SHAP) methods were utilized to interpret feature effects in heavy precipitation forecasting. The results indicated that precipitable water vapor(PWV), cloud, relative humidity, and seasonal and diurnal variables had more significant effects on the model output as individual features. Furthermore, the collective influence of derivatives from PWV and meteorological parameters(e.g., temperature, relative humidity, pressure and dew point temperature)showed a significant enhancement over their individual impacts, indicating synergistic interactions among these predictors.Applying explainable artificial intelligence(XAI) to ML models helps understand how models utilize features for forecasting, enhances the reliability of forecasts, and guides feature selection and the mitigation of overfitting phenomena.展开更多
Effective water distribution and transparency are threatened with being outrightly undermined unless the good name of urban infrastructure is maintained.With improved control systems in place to check leakage,variabil...Effective water distribution and transparency are threatened with being outrightly undermined unless the good name of urban infrastructure is maintained.With improved control systems in place to check leakage,variability of pressure,and conscientiousness of energy,issues that previously went unnoticed are now becoming recognized.This paper presents a grandiose hybrid framework that combines Multi-Agent Deep Reinforcement Learning(MADRL)with Shapley Additive Explanations(SHAP)-based Explainable AI(XAI)for adaptive and interpretable water resource management.In the methodology,the agents perform decentralized learning of the control policies for the pumps and valves based on the real-time network states,while also providing human-understandable explanations of the agents’decisions,using SHAP.This framework has been validated on five very diverse datasets,three of which are real-world scenarios involving actual water consumption from NYC and Alicante,with the other two being simulationbased standards such as LeakDB and the Water Distribution System Anomaly(WDSA)network.Empirical results demonstrate that the MADRL SHAP hybrid system reduces water loss by up to 32%,improves energy efficiency by+up to 25%,and maintains pressure stability between 91%and 93%,thereby outperforming the traditional rule-based control,single-agent DRL(Deep Reinforcement Learning),and XGBoost SHAP baselines.Furthermore,SHAP-based+interpretation brings transparency to the proposed model,with the average explanation consistency for all prediction models reaching 88%,thus further reinforcing the trustworthiness of the system on which the decision-making is based and empowering the utility operators to derive actionable insights from the model.The proposed framework addresses the critical challenges of smart water distribution.展开更多
With the advancements in artificial intelligence(AI)technology,attackers are increasingly using sophisticated techniques,including ChatGPT.Endpoint Detection&Response(EDR)is a system that detects and responds to s...With the advancements in artificial intelligence(AI)technology,attackers are increasingly using sophisticated techniques,including ChatGPT.Endpoint Detection&Response(EDR)is a system that detects and responds to strange activities or security threats occurring on computers or endpoint devices within an organization.Unlike traditional antivirus software,EDR is more about responding to a threat after it has already occurred than blocking it.This study aims to overcome challenges in security control,such as increased log size,emerging security threats,and technical demands faced by control staff.Previous studies have focused on AI detection models,emphasizing detection rates and model performance.However,the underlying reasons behind the detection results were often insufficiently understood,leading to varying outcomes based on the learning model.Additionally,the presence of both structured or unstructured logs,the growth in new security threats,and increasing technical disparities among control staff members pose further challenges for effective security control.This study proposed to improve the problems of the existing EDR system and overcome the limitations of security control.This study analyzed data during the preprocessing stage to identify potential threat factors that influence the detection process and its outcomes.Additionally,eleven commonly-used machine learning(ML)models for malware detection in XAI were tested,with the five models showing the highest performance selected for further analysis.Explainable AI(XAI)techniques are employed to assess the impact of preprocessing on the learning process outcomes.To ensure objectivity and versatility in the analysis,five widely recognized datasets were used.Additionally,eleven commonly-used machine learning models for malware detection in XAI were tested with the five models showing the highest performance selected for further analysis.The results indicate that eXtreme Gradient Boosting(XGBoost)model outperformed others.Moreover,the study conducts an in-depth analysis of the preprocessing phase,tracing backward from the detection result to infer potential threats and classify the primary variables influencing the model’s prediction.This analysis includes the application of SHapley Additive exPlanations(SHAP),an XAI result,which provides insight into the influence of specific features on detection outcomes,and suggests potential breaches by identifying common parameters in malware through file backtracking and providing weights.This study also proposed a counter-detection analysis process to overcome the limitations of existing Deep Learning outcomes,understand the decision-making process of AI,and enhance reliability.These contributions are expected to significantly enhance EDR systems and address existing limitations in security control.展开更多
People learn causal relations since childhood using counterfactual reasoning. Counterfactual reasoning uses counterfactual examples which take the form of “what if this has happened differently”. Counterfactual exam...People learn causal relations since childhood using counterfactual reasoning. Counterfactual reasoning uses counterfactual examples which take the form of “what if this has happened differently”. Counterfactual examples are also the basis of counterfactual explanation in explainable artificial intelligence (XAI). However, a framework that relies solely on optimization algorithms to find and present counterfactual samples cannot help users gain a deeper understanding of the system. Without a way to verify their understanding, the users can even be misled by such explanations. Such limitations can be overcome through an interactive and iterative framework that allows the users to explore their desired “what-if” scenarios. The purpose of our research is to develop such a framework. In this paper, we present our “what-if” XAI framework (WiXAI), which visualizes the artificial intelligence (AI) classification model from the perspective of the user’s sample and guides their “what-if” exploration. We also formulated how to use the WiXAI framework to generate counterfactuals and understand the feature-feature and feature-output relations in-depth for a local sample. These relations help move the users toward causal understanding.展开更多
The use of Explainable Artificial Intelligence(XAI)models becomes increasingly important for making decisions in smart healthcare environments.It is to make sure that decisions are based on trustworthy algorithms and ...The use of Explainable Artificial Intelligence(XAI)models becomes increasingly important for making decisions in smart healthcare environments.It is to make sure that decisions are based on trustworthy algorithms and that healthcare workers understand the decisions made by these algorithms.These models can potentially enhance interpretability and explainability in decision-making processes that rely on artificial intelligence.Nevertheless,the intricate nature of the healthcare field necessitates the utilization of sophisticated models to classify cancer images.This research presents an advanced investigation of XAI models to classify cancer images.It describes the different levels of explainability and interpretability associated with XAI models and the challenges faced in deploying them in healthcare applications.In addition,this study proposes a novel framework for cancer image classification that incorporates XAI models with deep learning and advanced medical imaging techniques.The proposed model integrates several techniques,including end-to-end explainable evaluation,rule-based explanation,and useradaptive explanation.The proposed XAI reaches 97.72%accuracy,90.72%precision,93.72%recall,96.72%F1-score,9.55%FDR,9.66%FOR,and 91.18%DOR.It will discuss the potential applications of the proposed XAI models in the smart healthcare environment.It will help ensure trust and accountability in AI-based decisions,which is essential for achieving a safe and reliable smart healthcare environment.展开更多
Electricity markets are highly complex,involving lots of interactions and complex dependencies that make it hard to understand the inner workings of the market and what is driving prices.Econometric methods have been ...Electricity markets are highly complex,involving lots of interactions and complex dependencies that make it hard to understand the inner workings of the market and what is driving prices.Econometric methods have been developed for this,white-box models,however,they are not as powerful as deep neural network models(DNN).In this paper,we use a DNN to forecast the price and then use XAI methods to understand the factors driving the price dynamics in the market.The objective is to increase our understanding of how different electricity markets work.To do that,we apply explainable methods such as SHAP and Gradient,combined with visual techniques like heatmaps(saliency maps)to analyse the behaviour and contributions of various features across five electricity markets.We introduce the novel concepts of SSHAP values and SSHAP lines to enhance the complex representation of high-dimensional tabular models.展开更多
The advancement of artificial intelligence(AI)in material design and engineering has led to significant improvements in predictive modeling of material properties.However,the lack of interpretability in machine learni...The advancement of artificial intelligence(AI)in material design and engineering has led to significant improvements in predictive modeling of material properties.However,the lack of interpretability in machine learning(ML)-based material informatics presents a major barrier to its practical adoption.This study proposes a novel quantitative computational framework that integrates ML models with explainable artificial intelligence(XAI)techniques to enhance both predictive accuracy and interpretability in material property prediction.The framework systematically incorporates a structured pipeline,including data processing,feature selection,model training,performance evaluation,explainability analysis,and real-world deployment.It is validated through a representative case study on the prediction of high-performance concrete(HPC)compressive strength,utilizing a comparative analysis of ML models such as Random Forest,XGBoost,Support Vector Regression(SVR),and Deep Neural Networks(DNNs).The results demonstrate that XGBoost achieves the highest predictive performance(R^(2)=0.918),while SHAP(Shapley Additive Explanations)and LIME(Local Interpretable Model-Agnostic Explanations)provide detailed insights into feature importance and material interactions.Additionally,the deployment of the trained model as a cloud-based Flask-Gunicorn API enables real-time inference,ensuring its scalability and accessibility for industrial and research applications.The proposed framework addresses key limitations of existing ML approaches by integrating advanced explainability techniques,systematically handling nonlinear feature interactions,and providing a scalable deployment strategy.This study contributes to the development of interpretable and deployable AI-driven material informatics,bridging the gap between data-driven predictions and fundamental material science principles.展开更多
文摘Alzheimer’s disease(AD)is a significant challenge in modern healthcare,with early detection and accurate staging remaining critical priorities for effective intervention.While Deep Learning(DL)approaches have shown promise in AD diagnosis,existing methods often struggle with the issues of precision,interpretability,and class imbalance.This study presents a novel framework that integrates DL with several eXplainable Artificial Intelligence(XAI)techniques,in particular attention mechanisms,Gradient-Weighted Class Activation Mapping(Grad-CAM),and Local Interpretable Model-Agnostic Explanations(LIME),to improve bothmodel interpretability and feature selection.The study evaluates four different DL architectures(ResMLP,VGG16,Xception,and Convolutional Neural Network(CNN)with attention mechanism)on a balanced dataset of 3714 MRI brain scans from patients aged 70 and older.The proposed CNN with attention model achieved superior performance,demonstrating 99.18%accuracy on the primary dataset and 96.64% accuracy on the ADNI dataset,significantly advancing the state-of-the-art in AD classification.The ability of the framework to provide comprehensive,interpretable results through multiple visualization techniques while maintaining high classification accuracy represents a significant advancement in the computational diagnosis of AD,potentially enabling more accurate and earlier intervention in clinical settings.
基金funded by the National Science and Technology Major Project under Grant No.2024ZD0532000 and Grant No.2024ZD0532002the National Natural Science Foundation of China under Grant No.62173318+2 种基金the Shenzhen Basic Research Program under Grant No.JCYJ20250604182831042the Key Laboratory of Biomedical Imaging Science and System,Chinese Academy of Sciencesthe Alliance of International Science Organization(ANSO)under Grant No.2021A8017729010.
文摘Diabetes imposes a substantial burden on global healthcare systems.Worldwide,nearly half of individuals with diabetes remain undiagnosed,while conventional diagnostic techniques are often invasive,painful,and expensive.In this study,we propose a noninvasive approach for diabetes detection using photoplethysmography(PPG),which is widely integrated into modern wearable devices.First,we derived velocity plethysmography(VPG)and acceleration plethysmography(APG)signals from PPG to construct multi-channel waveform representations.Second,we introduced a novel multiset spatiotemporal feature fusion framework that integrates hand-crafted temporal,statistical,and nonlinear features with recursive feature elimination and deep feature extraction using a one-dimensional statistical convolutional neural network(1DSCNN).Finally,we developed an interpretable diabetes detection method based on XGBoost,with explainable artificial intelligence(XAI)techniques.Specifically,SHapley Additive exPlanations(SHAP)and Local InterpretableModel-agnostic Explanations(LIME)were employed to identify and interpret potential digital biomarkers associated with diabetes.To validate the proposed method,we extended the publicly available Guilin People’s Hospital dataset by incorporating in-house clinical data from ten subjects,thereby enhancing data diversity.A subject-independent cross-validation strategy was applied to ensure that the testing subjects remained independent of the training data for robust generalization.Compared with existing state-of-the-art methods,our approach achieved superior performance,with an area under the curve(AUC)of 80.5±15.9%,sensitivity of 77.2±7.5%,and specificity of 64.3±18.2%.These results demonstrate that the proposed approach provides a noninvasive,interpretable,and accessible solution for diabetes detection using PPG signals.
基金Science and Technology Development Fund of Macao Special Administrative Region (0009/2024/RIB1)Guangdong Major Project of Basic and Applied Basic Research Foundation(2020B0301030004)。
文摘The inherent black-box nature of machine learning(ML) models limits their interpretability and broader application in heavy precipitation forecasting. Evaluating the reliability of these models involves analyzing the link between predictions and predictors. In this study, ERA5 reanalysis data, CERES satellite observations, and ground-based meteorological observatories were utilized to compile more comprehensive multi-type predictors for developing a Bayesian optimized XGBoost model for the nowcasting of heavy precipitation in the Guangdong-Hong Kong-Macao Greater Bay Area during the pre-summer rainy season. A comparison of model performance with different combinations of input features and classical machine learning algorithms demonstrated that the Bayesian optimized XGBoost model achieved the best overall performance, with an average Critical Success Index of 68.30%. Permutation Importance(PI) and shapley Additive Explanations(SHAP) methods were utilized to interpret feature effects in heavy precipitation forecasting. The results indicated that precipitable water vapor(PWV), cloud, relative humidity, and seasonal and diurnal variables had more significant effects on the model output as individual features. Furthermore, the collective influence of derivatives from PWV and meteorological parameters(e.g., temperature, relative humidity, pressure and dew point temperature)showed a significant enhancement over their individual impacts, indicating synergistic interactions among these predictors.Applying explainable artificial intelligence(XAI) to ML models helps understand how models utilize features for forecasting, enhances the reliability of forecasts, and guides feature selection and the mitigation of overfitting phenomena.
基金supported via funding from Prince sattam bin Abdulaziz University project number(PSAU/2025/R/1446).
文摘Effective water distribution and transparency are threatened with being outrightly undermined unless the good name of urban infrastructure is maintained.With improved control systems in place to check leakage,variability of pressure,and conscientiousness of energy,issues that previously went unnoticed are now becoming recognized.This paper presents a grandiose hybrid framework that combines Multi-Agent Deep Reinforcement Learning(MADRL)with Shapley Additive Explanations(SHAP)-based Explainable AI(XAI)for adaptive and interpretable water resource management.In the methodology,the agents perform decentralized learning of the control policies for the pumps and valves based on the real-time network states,while also providing human-understandable explanations of the agents’decisions,using SHAP.This framework has been validated on five very diverse datasets,three of which are real-world scenarios involving actual water consumption from NYC and Alicante,with the other two being simulationbased standards such as LeakDB and the Water Distribution System Anomaly(WDSA)network.Empirical results demonstrate that the MADRL SHAP hybrid system reduces water loss by up to 32%,improves energy efficiency by+up to 25%,and maintains pressure stability between 91%and 93%,thereby outperforming the traditional rule-based control,single-agent DRL(Deep Reinforcement Learning),and XGBoost SHAP baselines.Furthermore,SHAP-based+interpretation brings transparency to the proposed model,with the average explanation consistency for all prediction models reaching 88%,thus further reinforcing the trustworthiness of the system on which the decision-making is based and empowering the utility operators to derive actionable insights from the model.The proposed framework addresses the critical challenges of smart water distribution.
基金supported by Innovative Human Resource Development for Local Intellectualization program through the Institute of Information&Communications Technology Planning&Evaluation(IITP)grant funded by the Korea government(MSIT)(IITP-2024-RS-2022-00156287,50%)supported by the MSIT(Ministry of Science and ICT),Republic of Korea,under the Convergence Security Core Talent Training Business Support Program(IITP-2024-RS-2022-II221203,50%)supervised by the IITP(Institute of Information&Communications Technology Planning&Evaluation).
文摘With the advancements in artificial intelligence(AI)technology,attackers are increasingly using sophisticated techniques,including ChatGPT.Endpoint Detection&Response(EDR)is a system that detects and responds to strange activities or security threats occurring on computers or endpoint devices within an organization.Unlike traditional antivirus software,EDR is more about responding to a threat after it has already occurred than blocking it.This study aims to overcome challenges in security control,such as increased log size,emerging security threats,and technical demands faced by control staff.Previous studies have focused on AI detection models,emphasizing detection rates and model performance.However,the underlying reasons behind the detection results were often insufficiently understood,leading to varying outcomes based on the learning model.Additionally,the presence of both structured or unstructured logs,the growth in new security threats,and increasing technical disparities among control staff members pose further challenges for effective security control.This study proposed to improve the problems of the existing EDR system and overcome the limitations of security control.This study analyzed data during the preprocessing stage to identify potential threat factors that influence the detection process and its outcomes.Additionally,eleven commonly-used machine learning(ML)models for malware detection in XAI were tested,with the five models showing the highest performance selected for further analysis.Explainable AI(XAI)techniques are employed to assess the impact of preprocessing on the learning process outcomes.To ensure objectivity and versatility in the analysis,five widely recognized datasets were used.Additionally,eleven commonly-used machine learning models for malware detection in XAI were tested with the five models showing the highest performance selected for further analysis.The results indicate that eXtreme Gradient Boosting(XGBoost)model outperformed others.Moreover,the study conducts an in-depth analysis of the preprocessing phase,tracing backward from the detection result to infer potential threats and classify the primary variables influencing the model’s prediction.This analysis includes the application of SHapley Additive exPlanations(SHAP),an XAI result,which provides insight into the influence of specific features on detection outcomes,and suggests potential breaches by identifying common parameters in malware through file backtracking and providing weights.This study also proposed a counter-detection analysis process to overcome the limitations of existing Deep Learning outcomes,understand the decision-making process of AI,and enhance reliability.These contributions are expected to significantly enhance EDR systems and address existing limitations in security control.
文摘People learn causal relations since childhood using counterfactual reasoning. Counterfactual reasoning uses counterfactual examples which take the form of “what if this has happened differently”. Counterfactual examples are also the basis of counterfactual explanation in explainable artificial intelligence (XAI). However, a framework that relies solely on optimization algorithms to find and present counterfactual samples cannot help users gain a deeper understanding of the system. Without a way to verify their understanding, the users can even be misled by such explanations. Such limitations can be overcome through an interactive and iterative framework that allows the users to explore their desired “what-if” scenarios. The purpose of our research is to develop such a framework. In this paper, we present our “what-if” XAI framework (WiXAI), which visualizes the artificial intelligence (AI) classification model from the perspective of the user’s sample and guides their “what-if” exploration. We also formulated how to use the WiXAI framework to generate counterfactuals and understand the feature-feature and feature-output relations in-depth for a local sample. These relations help move the users toward causal understanding.
基金supported by theCONAHCYT(Consejo Nacional deHumanidades,Ciencias y Tecnologias).
文摘The use of Explainable Artificial Intelligence(XAI)models becomes increasingly important for making decisions in smart healthcare environments.It is to make sure that decisions are based on trustworthy algorithms and that healthcare workers understand the decisions made by these algorithms.These models can potentially enhance interpretability and explainability in decision-making processes that rely on artificial intelligence.Nevertheless,the intricate nature of the healthcare field necessitates the utilization of sophisticated models to classify cancer images.This research presents an advanced investigation of XAI models to classify cancer images.It describes the different levels of explainability and interpretability associated with XAI models and the challenges faced in deploying them in healthcare applications.In addition,this study proposes a novel framework for cancer image classification that incorporates XAI models with deep learning and advanced medical imaging techniques.The proposed model integrates several techniques,including end-to-end explainable evaluation,rule-based explanation,and useradaptive explanation.The proposed XAI reaches 97.72%accuracy,90.72%precision,93.72%recall,96.72%F1-score,9.55%FDR,9.66%FOR,and 91.18%DOR.It will discuss the potential applications of the proposed XAI models in the smart healthcare environment.It will help ensure trust and accountability in AI-based decisions,which is essential for achieving a safe and reliable smart healthcare environment.
基金Tsupported by EDF Energy R&D UK Centre Limited and EPSRC under Grant EP/V519625/1.
文摘Electricity markets are highly complex,involving lots of interactions and complex dependencies that make it hard to understand the inner workings of the market and what is driving prices.Econometric methods have been developed for this,white-box models,however,they are not as powerful as deep neural network models(DNN).In this paper,we use a DNN to forecast the price and then use XAI methods to understand the factors driving the price dynamics in the market.The objective is to increase our understanding of how different electricity markets work.To do that,we apply explainable methods such as SHAP and Gradient,combined with visual techniques like heatmaps(saliency maps)to analyse the behaviour and contributions of various features across five electricity markets.We introduce the novel concepts of SSHAP values and SSHAP lines to enhance the complex representation of high-dimensional tabular models.
基金supported by the J.Gustaf Richert Stiftelse(2023-00884)Energimyndigheten(P2021-00248)+3 种基金Svenska Forskningsrådet Formas(2022-01475)Kungl.Skogs-och Lantbruksakademien(GFS2023-0131BYG2023-0007GFS2024-0155)Royal Swedish Academy of Forestry and Agriculture(KSLA:GFS2023-0131,BYG2023-0007,GFS2024-0155)Anna and Nils Håkansson's Foundation(nhbidr24-6).
文摘The advancement of artificial intelligence(AI)in material design and engineering has led to significant improvements in predictive modeling of material properties.However,the lack of interpretability in machine learning(ML)-based material informatics presents a major barrier to its practical adoption.This study proposes a novel quantitative computational framework that integrates ML models with explainable artificial intelligence(XAI)techniques to enhance both predictive accuracy and interpretability in material property prediction.The framework systematically incorporates a structured pipeline,including data processing,feature selection,model training,performance evaluation,explainability analysis,and real-world deployment.It is validated through a representative case study on the prediction of high-performance concrete(HPC)compressive strength,utilizing a comparative analysis of ML models such as Random Forest,XGBoost,Support Vector Regression(SVR),and Deep Neural Networks(DNNs).The results demonstrate that XGBoost achieves the highest predictive performance(R^(2)=0.918),while SHAP(Shapley Additive Explanations)and LIME(Local Interpretable Model-Agnostic Explanations)provide detailed insights into feature importance and material interactions.Additionally,the deployment of the trained model as a cloud-based Flask-Gunicorn API enables real-time inference,ensuring its scalability and accessibility for industrial and research applications.The proposed framework addresses key limitations of existing ML approaches by integrating advanced explainability techniques,systematically handling nonlinear feature interactions,and providing a scalable deployment strategy.This study contributes to the development of interpretable and deployable AI-driven material informatics,bridging the gap between data-driven predictions and fundamental material science principles.
