The biological stabilization of soil using microbially induced carbonate precipitation(MICP)employs ureolytic bacteria to precipitate calcium carbonate(CaCO3),which binds soil particles,enhancing strength,stiffness,an...The biological stabilization of soil using microbially induced carbonate precipitation(MICP)employs ureolytic bacteria to precipitate calcium carbonate(CaCO3),which binds soil particles,enhancing strength,stiffness,and erosion resistance.The unconfinedcompressive strength(UCS),a key measure of soil strength,is critical in geotechnical engineering as it directly reflectsthe mechanical stability of treated soils.This study integrates explainable artificialintelligence(XAI)with geotechnical insights to model the UCS of MICP-treated sands.Using 517 experimental data points and a combination of various input variables—including median grain size(D50),coefficientof uniformity(Cu),void ratio(e),urea concentration(Mu),calcium concentration(Mc),optical density(OD)of bacterial solution,pH,and total injection volume(Vt)—fivemachine learning(ML)models,including eXtreme gradient boosting(XGBoost),Light gradient boosting machine(LightGBM),random forest(RF),gene expression programming(GEP),and multivariate adaptive regression splines(MARS),were developed and optimized.The ensemble models(XGBoost,LightGBM,and RF)were optimized using the Chernobyl disaster optimizer(CDO),a recently developed metaheuristic algorithm.Of these,LightGBM-CDO achieved the highest accuracy for UCS prediction.XAI techniques like feature importance analysis(FIA),SHapley additive exPlanations(SHAP),and partial dependence plots(PDPs)were also used to investigate the complex non-linear relationships between the input and output variables.The results obtained have demonstrated that the XAI-driven models can enhance the predictive accuracy and interpretability of MICP processes,offering a sustainable pathway for optimizing geotechnical applications.展开更多
Network attacks have become a critical issue in the internet security domain.Artificial intelligence technology-based detection methodologies have attracted attention;however,recent studies have struggled to adapt to ...Network attacks have become a critical issue in the internet security domain.Artificial intelligence technology-based detection methodologies have attracted attention;however,recent studies have struggled to adapt to changing attack patterns and complex network environments.In addition,it is difficult to explain the detection results logically using artificial intelligence.We propose a method for classifying network attacks using graph models to explain the detection results.First,we reconstruct the network packet data into a graphical structure.We then use a graph model to predict network attacks using edge classification.To explain the prediction results,we observed numerical changes by randomly masking and calculating the importance of neighbors,allowing us to extract significant subgraphs.Our experiments on six public datasets demonstrate superior performance with an average F1-score of 0.960 and accuracy of 0.964,outperforming traditional machine learning and other graph models.The visual representation of the extracted subgraphs highlights the neighboring nodes that have the greatest impact on the results,thus explaining detection.In conclusion,this study demonstrates that graph-based models are suitable for network attack detection in complex environments,and the importance of graph neighbors can be calculated to efficiently analyze the results.This approach can contribute to real-world network security analyses and provide a new direction in the field.展开更多
Most Convolutional Neural Network(CNN)interpretation techniques visualize only the dominant cues that the model relies on,but there is no guarantee that these represent all the evidence the model uses for classificati...Most Convolutional Neural Network(CNN)interpretation techniques visualize only the dominant cues that the model relies on,but there is no guarantee that these represent all the evidence the model uses for classification.This limitation becomes critical when hidden secondary cues—potentially more meaningful than the visualized ones—remain undiscovered.This study introduces CasCAM(Cascaded Class Activation Mapping)to address this fundamental limitation through counterfactual reasoning.By asking“if this dominant cue were absent,what other evidence would the model use?”,CasCAM progressively masks the most salient features and systematically uncovers the hierarchy of classification evidence hidden beneath them.Experimental results demonstrate that CasCAM effectively discovers the full spectrum of reasoning evidence and can be universally applied with nine existing interpretation methods.展开更多
Unconfined Compressive Strength(UCS)is a key parameter for the assessment of the stability and performance of stabilized soils,yet traditional laboratory testing is both time and resource intensive.In this study,an in...Unconfined Compressive Strength(UCS)is a key parameter for the assessment of the stability and performance of stabilized soils,yet traditional laboratory testing is both time and resource intensive.In this study,an interpretable machine learning approach to UCS prediction is presented,pairing five models(Random Forest(RF),Gradient Boosting(GB),Extreme Gradient Boosting(XGB),CatBoost,and K-Nearest Neighbors(KNN))with SHapley Additive exPlanations(SHAP)for enhanced interpretability and to guide feature removal.A complete dataset of 12 geotechnical and chemical parameters,i.e.,Atterberg limits,compaction properties,stabilizer chemistry,dosage,curing time,was used to train and test the models.R2,RMSE,MSE,and MAE were used to assess performance.Initial results with all 12 features indicated that boosting-based models(GB,XGB,CatBoost)exhibited the highest predictive accuracy(R^(2)=0.93)with satisfactory generalization on test data,followed by RF and KNN.SHAP analysis consistently picked CaO content,curing time,stabilizer dosage,and compaction parameters as the most important features,aligning with established soil stabilization mechanisms.Models were then re-trained on the top 8 and top 5 SHAP-ranked features.Interestingly,GB,XGB,and CatBoost maintained comparable accuracy with reduced input sets,while RF was moderately sensitive and KNN was somewhat better owing to reduced dimensionality.The findings confirm that feature reduction through SHAP enables cost-effective UCS prediction through the reduction of laboratory test requirements without significant accuracy loss.The suggested hybrid approach offers an explainable,interpretable,and cost-effective tool for geotechnical engineering practice.展开更多
Wildfires significantly disrupt the physical and hydrologic conditions of the environment,leading to vegetation loss and altered surface geo-material properties.These complex dynamics promote post-fire gully erosion,y...Wildfires significantly disrupt the physical and hydrologic conditions of the environment,leading to vegetation loss and altered surface geo-material properties.These complex dynamics promote post-fire gully erosion,yet the key conditioning factors(e.g.,topography,hydrology)remain insufficiently understood.This study proposes a novel artificial intelligence(AI)framework that integrates four machine learning(ML)models with Shapley Additive Explanations(SHAP)method,offering a hierarchical perspective from global to local on the dominant factors controlling gully distribution in wildfireaffected areas.In a case study of Xiangjiao catchment burned on March 28,2020,in Muli County in Sichuan Province of Southwest China,we derived 21 geoenvironmental factors to assess the susceptibility of post-fire gully erosion using logistic regression(LR),support vector machine(SVM),random forest(RF),and convolutional neural network(CNN)models.SHAP-based model interpretation revealed eight key conditioning factors:topographic position index(TPI),topographic wetness index(TWI),distance to stream,mean annual precipitation,differenced normalized burn ratio(d NBR),land use/cover,soil type,and distance to road.Comparative model evaluation demonstrated that reduced-variable models incorporating these dominant factors achieved accuracy comparable to that of the initial-variable models,with AUC values exceeding 0.868 across all ML algorithms.These findings provide critical insights into gully erosion behavior in wildfire-affected areas,supporting the decision-making process behind environmental management and hazard mitigation.展开更多
The increasing use of cloud-based devices has reached the critical point of cybersecurity and unwanted network traffic.Cloud environments pose significant challenges in maintaining privacy and security.Global approach...The increasing use of cloud-based devices has reached the critical point of cybersecurity and unwanted network traffic.Cloud environments pose significant challenges in maintaining privacy and security.Global approaches,such as IDS,have been developed to tackle these issues.However,most conventional Intrusion Detection System(IDS)models struggle with unseen cyberattacks and complex high-dimensional data.In fact,this paper introduces the idea of a novel distributed explainable and heterogeneous transformer-based intrusion detection system,named INTRUMER,which offers balanced accuracy,reliability,and security in cloud settings bymultiplemodulesworking together within it.The traffic captured from cloud devices is first passed to the TC&TM module in which the Falcon Optimization Algorithm optimizes the feature selection process,and Naie Bayes algorithm performs the classification of features.The selected features are classified further and are forwarded to the Heterogeneous Attention Transformer(HAT)module.In this module,the contextual interactions of the network traffic are taken into account to classify them as normal or malicious traffic.The classified results are further analyzed by the Explainable Prevention Module(XPM)to ensure trustworthiness by providing interpretable decisions.With the explanations fromthe classifier,emergency alarms are transmitted to nearby IDSmodules,servers,and underlying cloud devices for the enhancement of preventive measures.Extensive experiments on benchmark IDS datasets CICIDS 2017,Honeypots,and NSL-KDD were conducted to demonstrate the efficiency of the INTRUMER model in detecting network trafficwith high accuracy for different types.Theproposedmodel outperforms state-of-the-art approaches,obtaining better performance metrics:98.7%accuracy,97.5%precision,96.3%recall,and 97.8%F1-score.Such results validate the robustness and effectiveness of INTRUMER in securing diverse cloud environments against sophisticated cyber threats.展开更多
Problem:The integration of Artificial Intelligence(AI)into cybersecurity,while enhancing threat detection,is hampered by the“black box”nature of complex models,eroding trust,accountability,and regulatory compliance....Problem:The integration of Artificial Intelligence(AI)into cybersecurity,while enhancing threat detection,is hampered by the“black box”nature of complex models,eroding trust,accountability,and regulatory compliance.Explainable AI(XAI)aims to resolve this opacity but introduces a critical newvulnerability:the adversarial exploitation of model explanations themselves.Gap:Current research lacks a comprehensive synthesis of this dual role of XAI in cybersecurity—as both a tool for transparency and a potential attack vector.There is a pressing need to systematically analyze the trade-offs between interpretability and security,evaluate defense mechanisms,and outline a path for developing robust,next-generation XAI frameworks.Solution:This review provides a systematic examination of XAI techniques(e.g.,SHAP,LIME,Grad-CAM)and their applications in intrusion detection,malware analysis,and fraud prevention.It critically evaluates the security risks posed by XAI,including model inversion and explanation-guided evasion attacks,and assesses corresponding defense strategies such as adversarially robust training,differential privacy,and secure-XAI deployment patterns.Contribution:Theprimary contributions of this work are:(1)a comparative analysis of XAI methods tailored for cybersecurity contexts;(2)an identification of the critical trade-off betweenmodel interpretability and security robustness;(3)a synthesis of defense mechanisms to mitigate XAI-specific vulnerabilities;and(4)a forward-looking perspective proposing future research directions,including quantum-safe XAI,hybrid neuro-symbolic models,and the integration of XAI into Zero Trust Architectures.This review serves as a foundational resource for developing transparent,trustworthy,and resilient AI-driven cybersecurity systems.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
Although digital changes in power systems have added more ways to monitor and control them,these changes have also led to new cyber-attack risks,mainly from False Data Injection(FDI)attacks.If this happens,the sensors...Although digital changes in power systems have added more ways to monitor and control them,these changes have also led to new cyber-attack risks,mainly from False Data Injection(FDI)attacks.If this happens,the sensors and operations are compromised,which can lead to big problems,disruptions,failures and blackouts.In response to this challenge,this paper presents a reliable and innovative detection framework that leverages Bidirectional Long Short-Term Memory(Bi-LSTM)networks and employs explanatory methods from Artificial Intelligence(AI).Not only does the suggested architecture detect potential fraud with high accuracy,but it also makes its decisions transparent,enabling operators to take appropriate action.Themethod developed here utilizesmodel-free,interpretable tools to identify essential input elements,thereby making predictions more understandable and usable.Enhancing detection performance is made possible by correcting class imbalance using Synthetic Minority Over-sampling Technique(SMOTE)-based data balancing.Benchmark power system data confirms that the model functions correctly through detailed experiments.Experimental results showed that Bi-LSTM+Explainable AI(XAI)achieved an average accuracy of 94%,surpassing XGBoost(89%)and Bagging(84%),while ensuring explainability and a high level of robustness across various operating scenarios.By conducting an ablation study,we find that bidirectional recursive modeling and ReLU activation help improve generalization and model predictability.Additionally,examining model decisions through LIME enables us to identify which features are crucial for making smart grid operational decisions in real time.The research offers a practical and flexible approach for detecting FDI attacks,improving the security of cyber-physical systems,and facilitating the deployment of AI in energy infrastructure.展开更多
The diagnosis of brain tumors is an extended process that significantly depends on the expertise and skills of radiologists.The rise in patient numbers has substantially elevated the data processing volume,making conv...The diagnosis of brain tumors is an extended process that significantly depends on the expertise and skills of radiologists.The rise in patient numbers has substantially elevated the data processing volume,making conventional methods both costly and inefficient.Recently,Artificial Intelligence(AI)has gained prominence for developing automated systems that can accurately diagnose or segment brain tumors in a shorter time frame.Many researchers have examined various algorithms that provide both speed and accuracy in detecting and classifying brain tumors.This paper proposes a newmodel based on AI,called the Brain Tumor Detection(BTD)model,based on brain tumor Magnetic Resonance Images(MRIs).The proposed BTC comprises three main modules:(i)Image Processing Module(IPM),(ii)Patient Detection Module(PDM),and(iii)Explainable AI(XAI).In the first module(i.e.,IPM),the used dataset is preprocessed through two stages:feature extraction and feature selection.At first,the MRI is preprocessed,then the images are converted into a set of features using several feature extraction methods:gray level co-occurrencematrix,histogramof oriented gradient,local binary pattern,and Tamura feature.Next,the most effective features are selected fromthese features separately using ImprovedGrayWolfOptimization(IGWO).IGWOis a hybrid methodology that consists of the Filter Selection Step(FSS)using information gain ratio as an initial selection stage and Binary Gray Wolf Optimization(BGWO)to make the proposed method better at detecting tumors by further optimizing and improving the chosen features.Then,these features are fed to PDM using several classifiers,and the final decision is based on weighted majority voting.Finally,through Local Interpretable Model-agnostic Explanations(LIME)XAI,the interpretability and transparency in decision-making processes are provided.The experiments are performed on a publicly available Brain MRI dataset that consists of 98 normal cases and 154 abnormal cases.During the experiments,the dataset was divided into 70%(177 cases)for training and 30%(75 cases)for testing.The numerical findings demonstrate that the BTD model outperforms its competitors in terms of accuracy,precision,recall,and F-measure.It introduces 98.8%accuracy,97%precision,97.5%recall,and 97.2%F-measure.The results demonstrate the potential of the proposed model to revolutionize brain tumor diagnosis,contribute to better treatment strategies,and improve patient outcomes.展开更多
Breast cancer diagnosis relies heavily on many kinds of information from diverse sources—like mammogram images,ultrasound scans,patient records,and genetic tests—but most AI tools look at only one of these at a time...Breast cancer diagnosis relies heavily on many kinds of information from diverse sources—like mammogram images,ultrasound scans,patient records,and genetic tests—but most AI tools look at only one of these at a time,which limits their ability to produce accurate and comprehensive decisions.In recent years,multimodal learning has emerged,enabling the integration of heterogeneous data to improve performance and diagnostic accuracy.However,doctors cannot always see how or why these AI tools make their choices,which is a significant bottleneck in their reliability,along with adoption in clinical settings.Hence,people are adding explainable AI techniques that show the steps the model takes.This review investigates previous work that has employed multimodal learning and XAI for the diagnosis of breast cancer.It discusses the types of data,fusion techniques,and XAI models employed.It was done following the PRISMA guidelines and included studies from 2021 to April 2025.The literature search was performed systematically and resulted in 61 studies.The review highlights a gradual increase in current studies focusing on multimodal fusion and XAI,particularly in the years 2023–2024.It found that studies using multi-modal data fusion achieved the highest accuracy by 5%–10%on average compared to other studies that used single-modality data,an intermediate fusion strategy,and modern fusion techniques,such as cross attention,achieved the highest accuracy and best performance.The review also showed that SHAP,Grad-CAM,and LIME techniques are the most used in explaining breast cancer diagnostic models.There is a clear research shift toward integrating multimodal learning and XAI techniques into the breast cancer diagnostics field.However,several gaps were identified,including the scarcity of public multimodal datasets.Lack of a unified explainable framework in multimodal fusion systems,and lack of standardization in evaluating explanations.These limitations call for future research focused on building more shared datasets and integrating multimodal data and explainable AI techniques to improve decision-making and enhance transparency.展开更多
The relationship between the neighborhood environment and well-being is attracting increasingly attention from researchers and policymakers,as the goal of development has shift from economy to well-being.However,exist...The relationship between the neighborhood environment and well-being is attracting increasingly attention from researchers and policymakers,as the goal of development has shift from economy to well-being.However,existing literature predominantly adopts the utilitarian approach,understanding well-being as people’s feelings about their lives and viewing the neighborhood environment as resources that benefit well-being.The Capability Approach,a novel approach that conceptualize well-being as the freedoms to do or to be and regard environment as conversion factors that influence well-being,can offer new lens by incorporating human development in-to these topics.This paper proposes an alternative theoretical framework:well-being is conceptualized and measured by capability;neighborhood environment affects well-being by providing spatial services,functioning as environmental conversion factors,and serving as social conversion factors.We conducted a case study of Changshu City located in eastern China,utilizing multiple resource data,applying explainable artificial intelligence(XAI),namely eXtreme Gradient Boosting(XGBoost)and SHapley Additive exPlana-tions(SHAP).Our findings highlight the significance of viewing the neighborhood environment as a set of conversion factors,as it provides more explanatory power than providing spatial services.Compared to conventional research based on linear relationship as-sumption,our results demonstrate that the effects of neighborhood environment on well-being are non-linear,characterized by threshold effects and interaction effects.These insights are crucial for informing urban planning and public policy.This research enriches our un-derstanding of well-being,neighborhood environment,and their relationship as well as provides empirical evidence for the core concept of conversion factors in the capability approach.展开更多
Quality of Service(QoS)assurance in programmable IoT and 5G networks is increasingly threatened by cyberattacks such as Distributed Denial of Service(DDoS),spoofing,and botnet intrusions.This paper presents AutoSHARC,...Quality of Service(QoS)assurance in programmable IoT and 5G networks is increasingly threatened by cyberattacks such as Distributed Denial of Service(DDoS),spoofing,and botnet intrusions.This paper presents AutoSHARC,a feedback-driven,explainable intrusion detection framework that integrates Boruta and LightGBM–SHAP feature selection with a lightweight CNN–Attention–GRU classifier.AutoSHARC employs a two-stage feature selection pipeline to identify the most informative features from high-dimensional IoT traffic and reduces 46 features to 30 highly informative ones,followed by post-hoc SHAP-guided retraining to refine feature importance,forming a feedback loopwhere only the most impactful attributes are reused to retrain themodel.This iterative refinement reduces computational overhead,accelerates detection latency,and improves transparency.Evaluated on the CIC IoT 2023 dataset,AutoSHARC achieves 98.98%accuracy,98.9%F1-score,and strong robustness with a Matthews Correlation Coefficient of 0.98 and Cohen’s Kappa of 0.98.The final model contains only 531,272 trainable parameters with a compact 2 MB size,enabling real-time deployment on resource-constrained IoT nodes.By combining explainable AI with iterative feature refinement,AutoSHARC provides scalable and trustworthy intrusion detection while preserving key QoS indicators such as latency,throughput,and reliability.展开更多
In the field of precision healthcare,where accurate decision-making is paramount,this study underscores the indispensability of eXplainable Artificial Intelligence(XAI)in the context of epilepsy management within the ...In the field of precision healthcare,where accurate decision-making is paramount,this study underscores the indispensability of eXplainable Artificial Intelligence(XAI)in the context of epilepsy management within the Internet of Medical Things(IoMT).The methodology entails meticulous preprocessing,involving the application of a band-pass filter and epoch segmentation to optimize the quality of Electroencephalograph(EEG)data.The subsequent extraction of statistical features facilitates the differentiation between seizure and non-seizure patterns.The classification phase integrates Support Vector Machine(SVM),K-Nearest Neighbor(KNN),and Random Forest classifiers.Notably,SVM attains an accuracy of 97.26%,excelling in the precision,recall,specificity,and F1 score for identifying seizures and non-seizure instances.Conversely,KNN achieves an accuracy of 72.69%,accompanied by certain trade-offs.The Random Forest classifierstands out with a remarkable accuracy of 99.89%,coupled with an exceptional precision(99.73%),recall(100%),specificity(99.80%),and F1 score(99.86%),surpassing both SVM and KNN performances.XAI techniques,namely Local Interpretable ModelAgnostic Explanations(LIME)and SHapley Additive exPlanation(SHAP),enhance the system’s transparency.This combination of machine learning and XAI not only improves the reliability and accuracy of the seizure detection system but also enhances trust and interpretability.Healthcare professionals can leverage the identified important features and their dependencies to gain deeper insights into the decision-making process,aiding in informed diagnosis and treatment decisions for patients with epilepsy.展开更多
Early and accurate cancer diagnosis through medical imaging is crucial for guiding treatment and enhancing patient survival.However,many state-of-the-art deep learning(DL)methods remain opaque and lack clinical interp...Early and accurate cancer diagnosis through medical imaging is crucial for guiding treatment and enhancing patient survival.However,many state-of-the-art deep learning(DL)methods remain opaque and lack clinical interpretability.This paper presents an explainable artificial intelligence(XAI)framework that combines a fine-tuned Visual Geometry Group 16-layer network(VGG16)convolutional neural network with layer-wise relevance propagation(LRP)to deliver high-performance classification and transparent decision support.This approach is evaluated on the publicly available Kaggle kidney cancer imaging dataset,which comprises labeled cancerous and noncancerous kidney scans.The proposed model achieved 98.75%overall accuracy,with precision,recall,and F1-score each exceeding 98%on an independent test set.Crucially,LRP-derived heatmaps consistently localize anatomically and pathologically significant regions such as tumor margins in agreement with established clinical criteria.The proposed framework enhances clinician trust by delivering pixel-level justifications alongside state-of-the-art predictive performance.It facilitates informed decision-making,thereby addressing a key barrier to the clinical adoption of DL in oncology.展开更多
Predicting the health status of stroke patients at different stages of the disease is a critical clinical task.The onset and development of stroke are affected by an array of factors,encompassing genetic predispositio...Predicting the health status of stroke patients at different stages of the disease is a critical clinical task.The onset and development of stroke are affected by an array of factors,encompassing genetic predisposition,environmental exposure,unhealthy lifestyle habits,and existing medical conditions.Although existing machine learning-based methods for predicting stroke patients’health status have made significant progress,limitations remain in terms of prediction accuracy,model explainability,and system optimization.This paper proposes a multi-task learning approach based on Explainable Artificial Intelligence(XAI)for predicting the health status of stroke patients.First,we design a comprehensive multi-task learning framework that utilizes the task correlation of predicting various health status indicators in patients,enabling the parallel prediction of multiple health indicators.Second,we develop a multi-task Area Under Curve(AUC)optimization algorithm based on adaptive low-rank representation,which removes irrelevant information from the model structure to enhance the performance of multi-task AUC optimization.Additionally,the model’s explainability is analyzed through the stability analysis of SHAP values.Experimental results demonstrate that our approach outperforms comparison algorithms in key prognostic metrics F1 score and Efficiency.展开更多
In the evolving landscape of cyber threats,phishing attacks pose significant challenges,particularly through deceptive webpages designed to extract sensitive information under the guise of legitimacy.Conventional and ...In the evolving landscape of cyber threats,phishing attacks pose significant challenges,particularly through deceptive webpages designed to extract sensitive information under the guise of legitimacy.Conventional and machine learning(ML)-based detection systems struggle to detect phishing websites owing to their constantly changing tactics.Furthermore,newer phishing websites exhibit subtle and expertly concealed indicators that are not readily detectable.Hence,effective detection depends on identifying the most critical features.Traditional feature selection(FS)methods often struggle to enhance ML model performance and instead decrease it.To combat these issues,we propose an innovative method using explainable AI(XAI)to enhance FS in ML models and improve the identification of phishing websites.Specifically,we employ SHapley Additive exPlanations(SHAP)for global perspective and aggregated local interpretable model-agnostic explanations(LIME)to deter-mine specific localized patterns.The proposed SHAP and LIME-aggregated FS(SLA-FS)framework pinpoints the most informative features,enabling more precise,swift,and adaptable phishing detection.Applying this approach to an up-to-date web phishing dataset,we evaluate the performance of three ML models before and after FS to assess their effectiveness.Our findings reveal that random forest(RF),with an accuracy of 97.41%and XGBoost(XGB)at 97.21%significantly benefit from the SLA-FS framework,while k-nearest neighbors lags.Our framework increases the accuracy of RF and XGB by 0.65%and 0.41%,respectively,outperforming traditional filter or wrapper methods and any prior methods evaluated on this dataset,showcasing its potential.展开更多
文摘The biological stabilization of soil using microbially induced carbonate precipitation(MICP)employs ureolytic bacteria to precipitate calcium carbonate(CaCO3),which binds soil particles,enhancing strength,stiffness,and erosion resistance.The unconfinedcompressive strength(UCS),a key measure of soil strength,is critical in geotechnical engineering as it directly reflectsthe mechanical stability of treated soils.This study integrates explainable artificialintelligence(XAI)with geotechnical insights to model the UCS of MICP-treated sands.Using 517 experimental data points and a combination of various input variables—including median grain size(D50),coefficientof uniformity(Cu),void ratio(e),urea concentration(Mu),calcium concentration(Mc),optical density(OD)of bacterial solution,pH,and total injection volume(Vt)—fivemachine learning(ML)models,including eXtreme gradient boosting(XGBoost),Light gradient boosting machine(LightGBM),random forest(RF),gene expression programming(GEP),and multivariate adaptive regression splines(MARS),were developed and optimized.The ensemble models(XGBoost,LightGBM,and RF)were optimized using the Chernobyl disaster optimizer(CDO),a recently developed metaheuristic algorithm.Of these,LightGBM-CDO achieved the highest accuracy for UCS prediction.XAI techniques like feature importance analysis(FIA),SHapley additive exPlanations(SHAP),and partial dependence plots(PDPs)were also used to investigate the complex non-linear relationships between the input and output variables.The results obtained have demonstrated that the XAI-driven models can enhance the predictive accuracy and interpretability of MICP processes,offering a sustainable pathway for optimizing geotechnical applications.
基金supported by the MSIT(Ministry of Science and ICT),Republic of Korea,under the ICAN(ICT Challenge and Advanced Network of HRD)support program(IITP-2025-RS-2023-00259497)supervised by the IITP(Institute for Information&Communications Technology Planning&Evaluation)and was supported by Institute of Information&Communications Technology Planning&Evaluation(IITP)grant funded by the Republic of Korea government(MSIT)(No.IITP-2025-RS-2023-00254129+1 种基金Graduate School of Metaverse Convergence(Sungkyunkwan University))was supported by the Basic Science Research Program of the National Research Foundation(NRF)funded by the Republic of Korean government(MSIT)(No.RS-2024-00346737).
文摘Network attacks have become a critical issue in the internet security domain.Artificial intelligence technology-based detection methodologies have attracted attention;however,recent studies have struggled to adapt to changing attack patterns and complex network environments.In addition,it is difficult to explain the detection results logically using artificial intelligence.We propose a method for classifying network attacks using graph models to explain the detection results.First,we reconstruct the network packet data into a graphical structure.We then use a graph model to predict network attacks using edge classification.To explain the prediction results,we observed numerical changes by randomly masking and calculating the importance of neighbors,allowing us to extract significant subgraphs.Our experiments on six public datasets demonstrate superior performance with an average F1-score of 0.960 and accuracy of 0.964,outperforming traditional machine learning and other graph models.The visual representation of the extracted subgraphs highlights the neighboring nodes that have the greatest impact on the results,thus explaining detection.In conclusion,this study demonstrates that graph-based models are suitable for network attack detection in complex environments,and the importance of graph neighbors can be calculated to efficiently analyze the results.This approach can contribute to real-world network security analyses and provide a new direction in the field.
基金supported by the Basic Science Research Program through the National Research Foundation of Korea(NRF),funded by the Ministry of Education(RS-2023-00249743).
文摘Most Convolutional Neural Network(CNN)interpretation techniques visualize only the dominant cues that the model relies on,but there is no guarantee that these represent all the evidence the model uses for classification.This limitation becomes critical when hidden secondary cues—potentially more meaningful than the visualized ones—remain undiscovered.This study introduces CasCAM(Cascaded Class Activation Mapping)to address this fundamental limitation through counterfactual reasoning.By asking“if this dominant cue were absent,what other evidence would the model use?”,CasCAM progressively masks the most salient features and systematically uncovers the hierarchy of classification evidence hidden beneath them.Experimental results demonstrate that CasCAM effectively discovers the full spectrum of reasoning evidence and can be universally applied with nine existing interpretation methods.
文摘Unconfined Compressive Strength(UCS)is a key parameter for the assessment of the stability and performance of stabilized soils,yet traditional laboratory testing is both time and resource intensive.In this study,an interpretable machine learning approach to UCS prediction is presented,pairing five models(Random Forest(RF),Gradient Boosting(GB),Extreme Gradient Boosting(XGB),CatBoost,and K-Nearest Neighbors(KNN))with SHapley Additive exPlanations(SHAP)for enhanced interpretability and to guide feature removal.A complete dataset of 12 geotechnical and chemical parameters,i.e.,Atterberg limits,compaction properties,stabilizer chemistry,dosage,curing time,was used to train and test the models.R2,RMSE,MSE,and MAE were used to assess performance.Initial results with all 12 features indicated that boosting-based models(GB,XGB,CatBoost)exhibited the highest predictive accuracy(R^(2)=0.93)with satisfactory generalization on test data,followed by RF and KNN.SHAP analysis consistently picked CaO content,curing time,stabilizer dosage,and compaction parameters as the most important features,aligning with established soil stabilization mechanisms.Models were then re-trained on the top 8 and top 5 SHAP-ranked features.Interestingly,GB,XGB,and CatBoost maintained comparable accuracy with reduced input sets,while RF was moderately sensitive and KNN was somewhat better owing to reduced dimensionality.The findings confirm that feature reduction through SHAP enables cost-effective UCS prediction through the reduction of laboratory test requirements without significant accuracy loss.The suggested hybrid approach offers an explainable,interpretable,and cost-effective tool for geotechnical engineering practice.
基金the National Natural Science Foundation of China(42377170,42407212)the National Funded Postdoctoral Researcher Program(GZB20230606)+3 种基金the Postdoctoral Research Foundation of China(2024M752679)the Sichuan Natural Science Foundation(2025ZNSFSC1205)the National Key R&D Program of China(2022YFC3005704)the Sichuan Province Science and Technology Support Program(2024NSFSC0100)。
文摘Wildfires significantly disrupt the physical and hydrologic conditions of the environment,leading to vegetation loss and altered surface geo-material properties.These complex dynamics promote post-fire gully erosion,yet the key conditioning factors(e.g.,topography,hydrology)remain insufficiently understood.This study proposes a novel artificial intelligence(AI)framework that integrates four machine learning(ML)models with Shapley Additive Explanations(SHAP)method,offering a hierarchical perspective from global to local on the dominant factors controlling gully distribution in wildfireaffected areas.In a case study of Xiangjiao catchment burned on March 28,2020,in Muli County in Sichuan Province of Southwest China,we derived 21 geoenvironmental factors to assess the susceptibility of post-fire gully erosion using logistic regression(LR),support vector machine(SVM),random forest(RF),and convolutional neural network(CNN)models.SHAP-based model interpretation revealed eight key conditioning factors:topographic position index(TPI),topographic wetness index(TWI),distance to stream,mean annual precipitation,differenced normalized burn ratio(d NBR),land use/cover,soil type,and distance to road.Comparative model evaluation demonstrated that reduced-variable models incorporating these dominant factors achieved accuracy comparable to that of the initial-variable models,with AUC values exceeding 0.868 across all ML algorithms.These findings provide critical insights into gully erosion behavior in wildfire-affected areas,supporting the decision-making process behind environmental management and hazard mitigation.
文摘The increasing use of cloud-based devices has reached the critical point of cybersecurity and unwanted network traffic.Cloud environments pose significant challenges in maintaining privacy and security.Global approaches,such as IDS,have been developed to tackle these issues.However,most conventional Intrusion Detection System(IDS)models struggle with unseen cyberattacks and complex high-dimensional data.In fact,this paper introduces the idea of a novel distributed explainable and heterogeneous transformer-based intrusion detection system,named INTRUMER,which offers balanced accuracy,reliability,and security in cloud settings bymultiplemodulesworking together within it.The traffic captured from cloud devices is first passed to the TC&TM module in which the Falcon Optimization Algorithm optimizes the feature selection process,and Naie Bayes algorithm performs the classification of features.The selected features are classified further and are forwarded to the Heterogeneous Attention Transformer(HAT)module.In this module,the contextual interactions of the network traffic are taken into account to classify them as normal or malicious traffic.The classified results are further analyzed by the Explainable Prevention Module(XPM)to ensure trustworthiness by providing interpretable decisions.With the explanations fromthe classifier,emergency alarms are transmitted to nearby IDSmodules,servers,and underlying cloud devices for the enhancement of preventive measures.Extensive experiments on benchmark IDS datasets CICIDS 2017,Honeypots,and NSL-KDD were conducted to demonstrate the efficiency of the INTRUMER model in detecting network trafficwith high accuracy for different types.Theproposedmodel outperforms state-of-the-art approaches,obtaining better performance metrics:98.7%accuracy,97.5%precision,96.3%recall,and 97.8%F1-score.Such results validate the robustness and effectiveness of INTRUMER in securing diverse cloud environments against sophisticated cyber threats.
基金funded by the Deanship of Graduate Studies and Scientific Research at Jouf University under grant No.(DGSSR-2025-02-01395).
文摘Problem:The integration of Artificial Intelligence(AI)into cybersecurity,while enhancing threat detection,is hampered by the“black box”nature of complex models,eroding trust,accountability,and regulatory compliance.Explainable AI(XAI)aims to resolve this opacity but introduces a critical newvulnerability:the adversarial exploitation of model explanations themselves.Gap:Current research lacks a comprehensive synthesis of this dual role of XAI in cybersecurity—as both a tool for transparency and a potential attack vector.There is a pressing need to systematically analyze the trade-offs between interpretability and security,evaluate defense mechanisms,and outline a path for developing robust,next-generation XAI frameworks.Solution:This review provides a systematic examination of XAI techniques(e.g.,SHAP,LIME,Grad-CAM)and their applications in intrusion detection,malware analysis,and fraud prevention.It critically evaluates the security risks posed by XAI,including model inversion and explanation-guided evasion attacks,and assesses corresponding defense strategies such as adversarially robust training,differential privacy,and secure-XAI deployment patterns.Contribution:Theprimary contributions of this work are:(1)a comparative analysis of XAI methods tailored for cybersecurity contexts;(2)an identification of the critical trade-off betweenmodel interpretability and security robustness;(3)a synthesis of defense mechanisms to mitigate XAI-specific vulnerabilities;and(4)a forward-looking perspective proposing future research directions,including quantum-safe XAI,hybrid neuro-symbolic models,and the integration of XAI into Zero Trust Architectures.This review serves as a foundational resource for developing transparent,trustworthy,and resilient AI-driven cybersecurity systems.
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the Deanship of Scientific Research and Libraries in Princess Nourah bint Abdulrahman University for funding this research work through the Research Group project,Grant No.(RG-1445-0064).
文摘Although digital changes in power systems have added more ways to monitor and control them,these changes have also led to new cyber-attack risks,mainly from False Data Injection(FDI)attacks.If this happens,the sensors and operations are compromised,which can lead to big problems,disruptions,failures and blackouts.In response to this challenge,this paper presents a reliable and innovative detection framework that leverages Bidirectional Long Short-Term Memory(Bi-LSTM)networks and employs explanatory methods from Artificial Intelligence(AI).Not only does the suggested architecture detect potential fraud with high accuracy,but it also makes its decisions transparent,enabling operators to take appropriate action.Themethod developed here utilizesmodel-free,interpretable tools to identify essential input elements,thereby making predictions more understandable and usable.Enhancing detection performance is made possible by correcting class imbalance using Synthetic Minority Over-sampling Technique(SMOTE)-based data balancing.Benchmark power system data confirms that the model functions correctly through detailed experiments.Experimental results showed that Bi-LSTM+Explainable AI(XAI)achieved an average accuracy of 94%,surpassing XGBoost(89%)and Bagging(84%),while ensuring explainability and a high level of robustness across various operating scenarios.By conducting an ablation study,we find that bidirectional recursive modeling and ReLU activation help improve generalization and model predictability.Additionally,examining model decisions through LIME enables us to identify which features are crucial for making smart grid operational decisions in real time.The research offers a practical and flexible approach for detecting FDI attacks,improving the security of cyber-physical systems,and facilitating the deployment of AI in energy infrastructure.
文摘The diagnosis of brain tumors is an extended process that significantly depends on the expertise and skills of radiologists.The rise in patient numbers has substantially elevated the data processing volume,making conventional methods both costly and inefficient.Recently,Artificial Intelligence(AI)has gained prominence for developing automated systems that can accurately diagnose or segment brain tumors in a shorter time frame.Many researchers have examined various algorithms that provide both speed and accuracy in detecting and classifying brain tumors.This paper proposes a newmodel based on AI,called the Brain Tumor Detection(BTD)model,based on brain tumor Magnetic Resonance Images(MRIs).The proposed BTC comprises three main modules:(i)Image Processing Module(IPM),(ii)Patient Detection Module(PDM),and(iii)Explainable AI(XAI).In the first module(i.e.,IPM),the used dataset is preprocessed through two stages:feature extraction and feature selection.At first,the MRI is preprocessed,then the images are converted into a set of features using several feature extraction methods:gray level co-occurrencematrix,histogramof oriented gradient,local binary pattern,and Tamura feature.Next,the most effective features are selected fromthese features separately using ImprovedGrayWolfOptimization(IGWO).IGWOis a hybrid methodology that consists of the Filter Selection Step(FSS)using information gain ratio as an initial selection stage and Binary Gray Wolf Optimization(BGWO)to make the proposed method better at detecting tumors by further optimizing and improving the chosen features.Then,these features are fed to PDM using several classifiers,and the final decision is based on weighted majority voting.Finally,through Local Interpretable Model-agnostic Explanations(LIME)XAI,the interpretability and transparency in decision-making processes are provided.The experiments are performed on a publicly available Brain MRI dataset that consists of 98 normal cases and 154 abnormal cases.During the experiments,the dataset was divided into 70%(177 cases)for training and 30%(75 cases)for testing.The numerical findings demonstrate that the BTD model outperforms its competitors in terms of accuracy,precision,recall,and F-measure.It introduces 98.8%accuracy,97%precision,97.5%recall,and 97.2%F-measure.The results demonstrate the potential of the proposed model to revolutionize brain tumor diagnosis,contribute to better treatment strategies,and improve patient outcomes.
基金supported by the Deanship of Scientific Research,King Saud University through the Vice Deanship of Scientific Research Chairs,Chair of Pervasive and Mobile Computing.
文摘Breast cancer diagnosis relies heavily on many kinds of information from diverse sources—like mammogram images,ultrasound scans,patient records,and genetic tests—but most AI tools look at only one of these at a time,which limits their ability to produce accurate and comprehensive decisions.In recent years,multimodal learning has emerged,enabling the integration of heterogeneous data to improve performance and diagnostic accuracy.However,doctors cannot always see how or why these AI tools make their choices,which is a significant bottleneck in their reliability,along with adoption in clinical settings.Hence,people are adding explainable AI techniques that show the steps the model takes.This review investigates previous work that has employed multimodal learning and XAI for the diagnosis of breast cancer.It discusses the types of data,fusion techniques,and XAI models employed.It was done following the PRISMA guidelines and included studies from 2021 to April 2025.The literature search was performed systematically and resulted in 61 studies.The review highlights a gradual increase in current studies focusing on multimodal fusion and XAI,particularly in the years 2023–2024.It found that studies using multi-modal data fusion achieved the highest accuracy by 5%–10%on average compared to other studies that used single-modality data,an intermediate fusion strategy,and modern fusion techniques,such as cross attention,achieved the highest accuracy and best performance.The review also showed that SHAP,Grad-CAM,and LIME techniques are the most used in explaining breast cancer diagnostic models.There is a clear research shift toward integrating multimodal learning and XAI techniques into the breast cancer diagnostics field.However,several gaps were identified,including the scarcity of public multimodal datasets.Lack of a unified explainable framework in multimodal fusion systems,and lack of standardization in evaluating explanations.These limitations call for future research focused on building more shared datasets and integrating multimodal data and explainable AI techniques to improve decision-making and enhance transparency.
基金Under the auspices of National Natural Science Foundation of China(No.42271230,42330510)。
文摘The relationship between the neighborhood environment and well-being is attracting increasingly attention from researchers and policymakers,as the goal of development has shift from economy to well-being.However,existing literature predominantly adopts the utilitarian approach,understanding well-being as people’s feelings about their lives and viewing the neighborhood environment as resources that benefit well-being.The Capability Approach,a novel approach that conceptualize well-being as the freedoms to do or to be and regard environment as conversion factors that influence well-being,can offer new lens by incorporating human development in-to these topics.This paper proposes an alternative theoretical framework:well-being is conceptualized and measured by capability;neighborhood environment affects well-being by providing spatial services,functioning as environmental conversion factors,and serving as social conversion factors.We conducted a case study of Changshu City located in eastern China,utilizing multiple resource data,applying explainable artificial intelligence(XAI),namely eXtreme Gradient Boosting(XGBoost)and SHapley Additive exPlana-tions(SHAP).Our findings highlight the significance of viewing the neighborhood environment as a set of conversion factors,as it provides more explanatory power than providing spatial services.Compared to conventional research based on linear relationship as-sumption,our results demonstrate that the effects of neighborhood environment on well-being are non-linear,characterized by threshold effects and interaction effects.These insights are crucial for informing urban planning and public policy.This research enriches our un-derstanding of well-being,neighborhood environment,and their relationship as well as provides empirical evidence for the core concept of conversion factors in the capability approach.
文摘Quality of Service(QoS)assurance in programmable IoT and 5G networks is increasingly threatened by cyberattacks such as Distributed Denial of Service(DDoS),spoofing,and botnet intrusions.This paper presents AutoSHARC,a feedback-driven,explainable intrusion detection framework that integrates Boruta and LightGBM–SHAP feature selection with a lightweight CNN–Attention–GRU classifier.AutoSHARC employs a two-stage feature selection pipeline to identify the most informative features from high-dimensional IoT traffic and reduces 46 features to 30 highly informative ones,followed by post-hoc SHAP-guided retraining to refine feature importance,forming a feedback loopwhere only the most impactful attributes are reused to retrain themodel.This iterative refinement reduces computational overhead,accelerates detection latency,and improves transparency.Evaluated on the CIC IoT 2023 dataset,AutoSHARC achieves 98.98%accuracy,98.9%F1-score,and strong robustness with a Matthews Correlation Coefficient of 0.98 and Cohen’s Kappa of 0.98.The final model contains only 531,272 trainable parameters with a compact 2 MB size,enabling real-time deployment on resource-constrained IoT nodes.By combining explainable AI with iterative feature refinement,AutoSHARC provides scalable and trustworthy intrusion detection while preserving key QoS indicators such as latency,throughput,and reliability.
文摘In the field of precision healthcare,where accurate decision-making is paramount,this study underscores the indispensability of eXplainable Artificial Intelligence(XAI)in the context of epilepsy management within the Internet of Medical Things(IoMT).The methodology entails meticulous preprocessing,involving the application of a band-pass filter and epoch segmentation to optimize the quality of Electroencephalograph(EEG)data.The subsequent extraction of statistical features facilitates the differentiation between seizure and non-seizure patterns.The classification phase integrates Support Vector Machine(SVM),K-Nearest Neighbor(KNN),and Random Forest classifiers.Notably,SVM attains an accuracy of 97.26%,excelling in the precision,recall,specificity,and F1 score for identifying seizures and non-seizure instances.Conversely,KNN achieves an accuracy of 72.69%,accompanied by certain trade-offs.The Random Forest classifierstands out with a remarkable accuracy of 99.89%,coupled with an exceptional precision(99.73%),recall(100%),specificity(99.80%),and F1 score(99.86%),surpassing both SVM and KNN performances.XAI techniques,namely Local Interpretable ModelAgnostic Explanations(LIME)and SHapley Additive exPlanation(SHAP),enhance the system’s transparency.This combination of machine learning and XAI not only improves the reliability and accuracy of the seizure detection system but also enhances trust and interpretability.Healthcare professionals can leverage the identified important features and their dependencies to gain deeper insights into the decision-making process,aiding in informed diagnosis and treatment decisions for patients with epilepsy.
基金supported through the Ongoing Research Funding Program(ORF-2025-498),King Saud University,Riyadh,Saudi Arabia.
文摘Early and accurate cancer diagnosis through medical imaging is crucial for guiding treatment and enhancing patient survival.However,many state-of-the-art deep learning(DL)methods remain opaque and lack clinical interpretability.This paper presents an explainable artificial intelligence(XAI)framework that combines a fine-tuned Visual Geometry Group 16-layer network(VGG16)convolutional neural network with layer-wise relevance propagation(LRP)to deliver high-performance classification and transparent decision support.This approach is evaluated on the publicly available Kaggle kidney cancer imaging dataset,which comprises labeled cancerous and noncancerous kidney scans.The proposed model achieved 98.75%overall accuracy,with precision,recall,and F1-score each exceeding 98%on an independent test set.Crucially,LRP-derived heatmaps consistently localize anatomically and pathologically significant regions such as tumor margins in agreement with established clinical criteria.The proposed framework enhances clinician trust by delivering pixel-level justifications alongside state-of-the-art predictive performance.It facilitates informed decision-making,thereby addressing a key barrier to the clinical adoption of DL in oncology.
基金funded by the Excellent Talent Training Funding Project in Dongcheng District,Beijing,with project number 2024-dchrcpyzz-9.
文摘Predicting the health status of stroke patients at different stages of the disease is a critical clinical task.The onset and development of stroke are affected by an array of factors,encompassing genetic predisposition,environmental exposure,unhealthy lifestyle habits,and existing medical conditions.Although existing machine learning-based methods for predicting stroke patients’health status have made significant progress,limitations remain in terms of prediction accuracy,model explainability,and system optimization.This paper proposes a multi-task learning approach based on Explainable Artificial Intelligence(XAI)for predicting the health status of stroke patients.First,we design a comprehensive multi-task learning framework that utilizes the task correlation of predicting various health status indicators in patients,enabling the parallel prediction of multiple health indicators.Second,we develop a multi-task Area Under Curve(AUC)optimization algorithm based on adaptive low-rank representation,which removes irrelevant information from the model structure to enhance the performance of multi-task AUC optimization.Additionally,the model’s explainability is analyzed through the stability analysis of SHAP values.Experimental results demonstrate that our approach outperforms comparison algorithms in key prognostic metrics F1 score and Efficiency.
文摘In the evolving landscape of cyber threats,phishing attacks pose significant challenges,particularly through deceptive webpages designed to extract sensitive information under the guise of legitimacy.Conventional and machine learning(ML)-based detection systems struggle to detect phishing websites owing to their constantly changing tactics.Furthermore,newer phishing websites exhibit subtle and expertly concealed indicators that are not readily detectable.Hence,effective detection depends on identifying the most critical features.Traditional feature selection(FS)methods often struggle to enhance ML model performance and instead decrease it.To combat these issues,we propose an innovative method using explainable AI(XAI)to enhance FS in ML models and improve the identification of phishing websites.Specifically,we employ SHapley Additive exPlanations(SHAP)for global perspective and aggregated local interpretable model-agnostic explanations(LIME)to deter-mine specific localized patterns.The proposed SHAP and LIME-aggregated FS(SLA-FS)framework pinpoints the most informative features,enabling more precise,swift,and adaptable phishing detection.Applying this approach to an up-to-date web phishing dataset,we evaluate the performance of three ML models before and after FS to assess their effectiveness.Our findings reveal that random forest(RF),with an accuracy of 97.41%and XGBoost(XGB)at 97.21%significantly benefit from the SLA-FS framework,while k-nearest neighbors lags.Our framework increases the accuracy of RF and XGB by 0.65%and 0.41%,respectively,outperforming traditional filter or wrapper methods and any prior methods evaluated on this dataset,showcasing its potential.
