Self-Explaining Autonomous Systems(SEAS)have emerged as a strategic frontier within Artificial Intelligence(AI),responding to growing demands for transparency and interpretability in autonomous decisionmaking.This stu...Self-Explaining Autonomous Systems(SEAS)have emerged as a strategic frontier within Artificial Intelligence(AI),responding to growing demands for transparency and interpretability in autonomous decisionmaking.This study presents a comprehensive bibliometric analysis of SEAS research published between 2020 and February 2025,drawing upon 1380 documents indexed in Scopus.The analysis applies co-citation mapping,keyword co-occurrence,and author collaboration networks using VOSviewer,MASHA,and Python to examine scientific production,intellectual structure,and global collaboration patterns.The results indicate a sustained annual growth rate of 41.38%,with an h-index of 57 and an average of 21.97 citations per document.A normalized citation rate was computed to address temporal bias,enabling balanced evaluation across publication cohorts.Thematic analysis reveals four consolidated research fronts:interpretability in machine learning,explainability in deep neural networks,transparency in generative models,and optimization strategies in autonomous control.Author co-citation analysis identifies four distinct research communities,and keyword evolution shows growing interdisciplinary links with medicine,cybersecurity,and industrial automation.The United States leads in scientific output and citation impact at the geographical level,while countries like India and China show high productivity with varied influence.However,international collaboration remains limited at 7.39%,reflecting a fragmented research landscape.As discussed in this study,SEAS research is expanding rapidly yet remains epistemologically dispersed,with uneven integration of ethical and human-centered perspectives.This work offers a structured and data-driven perspective on SEAS development,highlights key contributors and thematic trends,and outlines critical directions for advancing responsible and transparent autonomous systems.展开更多
The integration of machine learning(ML)into geohazard assessment has successfully instigated a paradigm shift,leading to the production of models that possess a level of predictive accuracy previously considered unatt...The integration of machine learning(ML)into geohazard assessment has successfully instigated a paradigm shift,leading to the production of models that possess a level of predictive accuracy previously considered unattainable.However,the black-box nature of these systems presents a significant barrier,hindering their operational adoption,regulatory approval,and full scientific validation.This paper provides a systematic review and synthesis of the emerging field of explainable artificial intelligence(XAI)as applied to geohazard science(GeoXAI),a domain that aims to resolve the long-standing trade-off between model performance and interpretability.A rigorous synthesis of 87 foundational studies is used to map the intellectual and methodological contours of this rapidly expanding field.The analysis reveals that current research efforts are concentrated predominantly on landslide and flood assessment.Methodologically,tree-based ensembles and deep learning models dominate the literature,with SHapley Additive exPlanations(SHAP)frequently adopted as the principal post-hoc explanation technique.More importantly,the review further documents how the role of XAI has shifted:rather than being used solely as a tool for interpreting models after training,it is increasingly integrated into the modeling cycle itself.Recent applications include its use in feature selection,adaptive sampling strategies,and model evaluation.The evidence also shows that GeoXAI extends beyond producing feature rankings.It reveals nonlinear thresholds and interaction effects that generate deeper mechanistic insights into hazard processes and mechanisms.Nevertheless,several key challenges remain unresolved within the field.These persistent issues are especially pronounced when considering the crucial necessity for interpretation stability,the demanding scholarly task of reliably distinguishing correlation from causation,and the development of appropriate methods for the treatment of complex spatio-temporal dynamics.展开更多
Log anomaly detection is essential for maintaining the reliability and security of large-scale networked systems.Most traditional techniques rely on log parsing in the reprocessing stage and utilize handcrafted featur...Log anomaly detection is essential for maintaining the reliability and security of large-scale networked systems.Most traditional techniques rely on log parsing in the reprocessing stage and utilize handcrafted features that limit their adaptability across various systems.In this study,we propose a hybrid model,BertGCN,that integrates BERT-based contextual embedding with Graph Convolutional Networks(GCNs)to identify anomalies in raw system logs,thereby eliminating the need for log parsing.TheBERT module captures semantic representations of log messages,while the GCN models the structural relationships among log entries through a text-based graph.This combination enables BertGCN to capture both the contextual and semantic characteristics of log data.BertGCN showed excellent performance on the HDFS and BGL datasets,demonstrating its effectiveness and resilience in detecting anomalies.Compared to multiple baselines,our proposed BertGCN showed improved precision,recall,and F1 scores.展开更多
Over the past decade,the landscape of cybersecurity has been increasingly shaped by the growing sophistication and frequency of malware attacks.Traditional detection techniques,while still in use,often fall short when...Over the past decade,the landscape of cybersecurity has been increasingly shaped by the growing sophistication and frequency of malware attacks.Traditional detection techniques,while still in use,often fall short when confronted with modern threats that use advanced evasion strategies.This systematic review critically examines recent developments in malware detection,with a particular emphasis on the role of artificial intelligence(AI)and machine learning(ML)in enhancing detection capabilities.Drawing on literature published between 2019 and 2025,this study reviews 105 peer-reviewed contributions from prominent digital libraries including IEEE Xplore,SpringerLink,ScienceDirect,and ACM Digital Library.In doing so,it explores the evolution of malware,evaluates detection methods,assesses the quality and limitations of widely used datasets,and identifies key challenges facing the field.Unlike existing surveys,this work offers a structured comparison of AI-driven frameworks and provides a detailed account of emerging techniques such as hybrid detection frameworks and image-based analysis.The findings indicate that AIbased models trained on diverse,high-quality datasets consistently outperform conventional methods,particularly when supported by feature engineering,explainable AI and a multi-faceted strategy.The review concludes by outlining future research directions,including the need for standardized datasets,enhanced adversarial robustness,and the integration of privacy-preserving mechanisms in malware detection systems.展开更多
The increasing number of interconnected devices and the incorporation of smart technology into contemporary healthcare systems have significantly raised the attack surface of cyber threats.The early detection of threa...The increasing number of interconnected devices and the incorporation of smart technology into contemporary healthcare systems have significantly raised the attack surface of cyber threats.The early detection of threats is both necessary and complex,yet these interconnected healthcare settings generate enormous amounts of heterogeneous data.Traditional Intrusion Detection Systems(IDS),which are generally centralized and machine learning-based,often fail to address the rapidly changing nature of cyberattacks and are challenged by ethical concerns related to patient data privacy.Moreover,traditional AI-driven IDS usually face challenges in handling large-scale,heterogeneous healthcare data while ensuring data privacy and operational efficiency.To address these issues,emerging technologies such as Big Data Analytics(BDA)and Federated Learning(FL)provide a hybrid framework for scalable,adaptive intrusion detection in IoT-driven healthcare systems.Big data techniques enable processing large-scale,highdimensional healthcare data,and FL can be used to train a model in a decentralized manner without transferring raw data,thereby maintaining privacy between institutions.This research proposes a privacy-preserving Federated Learning–based model that efficiently detects cyber threats in connected healthcare systems while ensuring distributed big data processing,privacy,and compliance with ethical regulations.To strengthen the reliability of the reported findings,the resultswere validated using cross-dataset testing and 95%confidence intervals derived frombootstrap analysis,confirming consistent performance across heterogeneous healthcare data distributions.This solution takes a significant step toward securing next-generation healthcare infrastructure by combining scalability,privacy,adaptability,and earlydetection capabilities.The proposed global model achieves a test accuracy of 99.93%±0.03(95%CI)and amiss-rate of only 0.07%±0.02,representing state-of-the-art performance in privacy-preserving intrusion detection.The proposed FL-driven IDS framework offers an efficient,privacy-preserving,and scalable solution for securing next-generation healthcare infrastructures by combining adaptability,early detection,and ethical data management.展开更多
Illicit web ecosystems,encompassing phishing,illegal online gambling,scam platforms,and malicious advertising,have rapidly expanded in scale and complexity,creating severe social,financial,and cybersecurity risks.Trad...Illicit web ecosystems,encompassing phishing,illegal online gambling,scam platforms,and malicious advertising,have rapidly expanded in scale and complexity,creating severe social,financial,and cybersecurity risks.Traditional rule-based and blacklist-driven detection approaches struggle to cope with polymorphic,multilingual,and adversarially manipulated threats,resulting in increasing demand for Artificial Intelligence(AI)-based solutions.This review provides a comprehensive synthesis of research on AI-driven threat detection for illicit web environments.It surveys detection models across multiple modalities,including text-based analysis of Uniform Resource Locator(URL)and HyperText Markup Language(HTML),vision-based recognition of webpage layouts and logos,graphbased modeling of domain and infrastructure relationships,and sequence modeling using transformer architectures.In addition,the paper examines system architectures,data collection and labeling pipelines,real-time detection frameworks,and widely used benchmark datasets,while also discussing their inherent limitations related to imbalance,representativeness,and reproducibility.The review highlights critical challenges such as evasion strategies,cross-lingual detection barriers,deployment latency,and explainability gaps.Furthermore,it identifies emerging research directions,including the use of Generative Adversarial Network(GAN)for threat simulation,few-shot and self-supervised learning for data-scarce environments,Explainable Artificial Intelligence(XAI)for transparency,and predictive AI for proactive threat forecasting.By integrating technical,legal,and societal perspectives,this survey offers a structured foundation for researchers and practitioners to design resilient,adaptive,and trustworthy AI-based defense systems against illicit web threats.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
With the ongoing digitalization and intelligence of power systems,there is an increasing reliance on large-scale data-driven intelligent technologies for tasks such as scheduling optimization and load forecasting.Neve...With the ongoing digitalization and intelligence of power systems,there is an increasing reliance on large-scale data-driven intelligent technologies for tasks such as scheduling optimization and load forecasting.Nevertheless,power data often contains sensitive information,making it a critical industry challenge to efficiently utilize this data while ensuring privacy.Traditional Federated Learning(FL)methods can mitigate data leakage by training models locally instead of transmitting raw data.Despite this,FL still has privacy concerns,especially gradient leakage,which might expose users’sensitive information.Therefore,integrating Differential Privacy(DP)techniques is essential for stronger privacy protection.Even so,the noise from DP may reduce the performance of federated learning models.To address this challenge,this paper presents an explainability-driven power data privacy federated learning framework.It incorporates DP technology and,based on model explainability,adaptively adjusts privacy budget allocation and model aggregation,thus balancing privacy protection and model performance.The key innovations of this paper are as follows:(1)We propose an explainability-driven power data privacy federated learning framework.(2)We detail a privacy budget allocation strategy:assigning budgets per training round by gradient effectiveness and at model granularity by layer importance.(3)We design a weighted aggregation strategy that considers the SHAP value and model accuracy for quality knowledge sharing.(4)Experiments show the proposed framework outperforms traditional methods in balancing privacy protection and model performance in power load forecasting tasks.展开更多
Fingerprint classification is a biometric method for crime prevention.For the successful completion of various tasks,such as official attendance,banking transactions,andmembership requirements,fingerprint classificati...Fingerprint classification is a biometric method for crime prevention.For the successful completion of various tasks,such as official attendance,banking transactions,andmembership requirements,fingerprint classification methods require improvement in terms of accuracy,speed,and the interpretability of non-linear demographic features.Researchers have introduced several CNN-based fingerprint classification models with improved accuracy,but these models often lack effective feature extractionmechanisms and complex multineural architectures.In addition,existing literature primarily focuses on gender classification rather than accurately,efficiently,and confidently classifying hands and fingers through the interpretability of prominent features.This research seeks to improve a compact,robust,explainable,and non-linear feature extraction-based CNN model for robust fingerprint pattern analysis and accurate yet efficient fingerprint classification.The proposed model(a)recognizes gender,hands,and fingers correctly through an advanced channel-wise attention-based feature extraction procedure,(b)accelerates the fingerprints identification process by applying an innovative fractional optimizer within a simple,but effective classification architecture,and(c)interprets prominent features through an explainable artificial intelligence technique.The encapsulated dependencies among distinct complex features are captured through a non-linear activation operation within a customized CNN model.The proposed fractionally optimized convolutional neural network(FOCNN)model demonstrates improved performance compared to some existing models,achieving high accuracies of 97.85%,99.10%,and 99.29%for finger,gender,and hand classification,respectively,utilizing the benchmark Sokoto Coventry Fingerprint Dataset.展开更多
Feature selection(FS)plays a crucial role in medical imaging by reducing dimensionality,improving computational efficiency,and enhancing diagnostic accuracy.Traditional FS techniques,including filter,wrapper,and embed...Feature selection(FS)plays a crucial role in medical imaging by reducing dimensionality,improving computational efficiency,and enhancing diagnostic accuracy.Traditional FS techniques,including filter,wrapper,and embedded methods,have been widely used but often struggle with high-dimensional and heterogeneous medical imaging data.Deep learning-based FS methods,particularly Convolutional Neural Networks(CNNs)and autoencoders,have demonstrated superior performance but lack interpretability.Hybrid approaches that combine classical and deep learning techniques have emerged as a promising solution,offering improved accuracy and explainability.Furthermore,integratingmulti-modal imaging data(e.g.,MagneticResonance Imaging(MRI),ComputedTomography(CT),Positron Emission Tomography(PET),and Ultrasound(US))poses additional challenges in FS,necessitating advanced feature fusion strategies.Multi-modal feature fusion combines information fromdifferent imagingmodalities to improve diagnostic accuracy.Recently,quantum computing has gained attention as a revolutionary approach for FS,providing the potential to handle high-dimensional medical data more efficiently.This systematic literature review comprehensively examines classical,Deep Learning(DL),hybrid,and quantum-based FS techniques inmedical imaging.Key outcomes include a structured taxonomy of FS methods,a critical evaluation of their performance across modalities,and identification of core challenges such as computational burden,interpretability,and ethical considerations.Future research directions—such as explainable AI(XAI),federated learning,and quantum-enhanced FS—are also emphasized to bridge the current gaps.This review provides actionable insights for developing scalable,interpretable,and clinically applicable FS methods in the evolving landscape of medical imaging.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 convergence of Software Defined Networking(SDN)in Internet of Vehicles(IoV)enables a flexible,programmable,and globally visible network control architecture across Road Side Units(RSUs),cloud servers,and automobil...The convergence of Software Defined Networking(SDN)in Internet of Vehicles(IoV)enables a flexible,programmable,and globally visible network control architecture across Road Side Units(RSUs),cloud servers,and automobiles.While this integration enhances scalability and safety,it also raises sophisticated cyberthreats,particularly Distributed Denial of Service(DDoS)attacks.Traditional rule-based anomaly detection methods often struggle to detectmodern low-and-slowDDoS patterns,thereby leading to higher false positives.To this end,this study proposes an explainable hybrid framework to detect DDoS attacks in SDN-enabled IoV(SDN-IoV).The hybrid framework utilizes a Residual Network(ResNet)to capture spatial correlations and a Bi-Long Short-Term Memory(BiLSTM)to capture both forward and backward temporal dependencies in high-dimensional input patterns.To ensure transparency and trustworthiness,themodel integrates the Explainable AI(XAI)technique,i.e.,SHapley Additive exPlanations(SHAP).SHAP highlights the contribution of each feature during the decision-making process,facilitating security analysts to understand the rationale behind the attack classification decision.The SDN-IoV environment is created in Mininet-WiFi and SUMO,and the hybrid model is trained on the CICDDoS2019 security dataset.The simulation results reveal the efficacy of the proposed model in terms of standard performance metrics compared to similar baseline methods.展开更多
Coordinate transformation models often fail to account for nonlinear and spatially dependent distortions,leading to significant residual errors in geospatial applications.Here,we propose a residual-based neural correc...Coordinate transformation models often fail to account for nonlinear and spatially dependent distortions,leading to significant residual errors in geospatial applications.Here,we propose a residual-based neural correction(RBNC)strategy,in which a neural network learns to model only the systematic distortions left by an initial geometric transformation.By focusing solely on residual patterns,RBNC reduces model complexity and improves performance,particularly in scenarios with sparse or structured control point configurations.We evaluate the method using both simulated datasets(with varying distortion intensities and sampling strategies)and real-world image georeferencing tasks.Compared with direct neural network coordinate converters and classical transformation models,RBNC delivers more accurate and stable results under challenging conditions,while maintaining comparable performance in ideal cases.These findings demonstrate the effectiveness of residual modelling as a light-weight and robust alternative for improving coordinate transformation accuracy.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
Objective:Deep learning is employed increasingly in Gastroenterology(GI)endoscopy computer-aided diagnostics for polyp segmentation and multi-class disease detection.In the real world,implementation requires high accu...Objective:Deep learning is employed increasingly in Gastroenterology(GI)endoscopy computer-aided diagnostics for polyp segmentation and multi-class disease detection.In the real world,implementation requires high accuracy,therapeutically relevant explanations,strong calibration,domain generalization,and efficiency.Current Convolutional Neural Network(CNN)and transformer models compromise border precision and global context,generate attention maps that fail to align with expert reasoning,deteriorate during cross-center changes,and exhibit inadequate calibration,hence diminishing clinical trust.Methods:HMA-DER is a hierarchical multi-attention architecture that uses dilation-enhanced residual blocks and an explainability-aware Cognitive Alignment Score(CAS)regularizer to directly align attribution maps with reasoning signals from experts.The framework has additions that make it more resilient and a way to test for accuracy,macro-averaged F1 score,Area Under the Receiver Operating Characteristic Curve(AUROC),calibration(Expected Calibration Error(ECE),Brier Score),explainability(CAS,insertion/deletion AUC),cross-dataset transfer,and throughput.Results:HMA-DER gets Dice Similarity Coefficient scores of 89.5%and 86.0%on Kvasir-SEG and CVC-ClinicDB,beating the strongest baseline by+1.9 and+1.7 points.It gets 86.4%and 85.3%macro-F1 and 94.0%and 93.4%AUROC on HyperKvasir and GastroVision,which is better than the baseline by+1.4/+1.6macro-F1 and+1.2/+1.1AUROC.Ablation study shows that hierarchical attention gives the highest(+3.0),followed by CAS regularization(+2–3),dilatation(+1.5–2.0),and residual connections(+2–3).Cross-dataset validation demonstrates competitive zero-shot transfer(e.g.,KS→CVC Dice 82.7%),whereas multi-dataset training diminishes the domain gap,yielding an 88.1%primary-metric average.HMA-DER’s mixed-precision inference can handle 155 pictures per second,which helps with calibration.Conclusion:HMA-DER strikes a compromise between accuracy,explainability,robustness,and efficiency for the use of reliable GI computer-aided diagnosis in real-world clinical settings.展开更多
Ovarian cancer(OC)is one of the leading causes of death related to gynecological cancer,with the main difficulty of its early diagnosis and a heterogeneous nature of tumor biomarkers.Machine learning(ML)has the potent...Ovarian cancer(OC)is one of the leading causes of death related to gynecological cancer,with the main difficulty of its early diagnosis and a heterogeneous nature of tumor biomarkers.Machine learning(ML)has the potential to process complex datasets and support decision-making in OC diagnosis.Nevertheless,traditional ML models tend to be biased,overfitting,noisy,and less generalized.Moreover,their black-box nature reduces interpretability and limits their practical clinical applicability.In this study,we introduce an explainable ensemble learning(EL)model,TreeX-Stack,based on a stacking architecture that employs tree-based learners such as Decision Tree(DT),Random Forest(RF),Gradient Boosting(GB),and Extreme Gradient Boosting(XGBoost)as base learners,and Logistic Regression(LR)as the meta-learner to enhance ovarian cancer(OC)diagnosis.Local Interpretable ModelAgnostic Explanations(LIME)are used to explain individual predictions,making the model outputs more clinically interpretable and applicable.The model is trained on the dataset that includes demographic information,blood test,general chemistry,and tumor markers.Extensive preprocessing includes handling missing data using iterative imputation with Bayesian Ridge and addressing multicollinearity by removing features with correlation coefficients above 0.7.Relevant features are then selected using the Boruta feature selection method.To obtain robust and unbiased performance estimates during hyperparameter tuning,nested cross-validation(CV)with grid search is employed,and all experiments are repeated five times to ensure statistical reliability.TreeX-Stack demonstrates excellent diagnostic performance,achieving an accuracy of 0.9027,a precision of 0.8673,a recall of 0.9391,and an F1-score of 0.9012.Feature-importance analyses using LIME and permutation importance highlight Human Epididymis Protein 4(HE4)as the most significant biomarker for OC.The combination of high predictive performance and interpretability makes TreeX-Stack a reliable tool for clinical decision support in OC diagnosis.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
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
Globally,diabetes and glaucoma account for a high number of people suffering from severe vision loss and blindness.To treat these vision disorders effectively,proper diagnosis must occur in a timely manner,and with co...Globally,diabetes and glaucoma account for a high number of people suffering from severe vision loss and blindness.To treat these vision disorders effectively,proper diagnosis must occur in a timely manner,and with conventional methods such as fundus photography,optical coherence tomography(OCT),and slit-lamp imaging,much depends on an expert’s interpretation of the images,making the systems very labor-intensive to operate.Moreover,clinical settings face difficulties with inter-observer variability and limited scalability with these diagnostic devices.To solve these problems,we have developed the Efficient Channel-Spatial Attention Network(ECSA-Net),a new deep learning-based methodology that integrates lightweight channel-and spatial-attention modules into a convolutional neural network.Ultimately,ECSA-Net improves the efficiency of computational resource use while enhancing discriminative feature extraction from retinal images.The ECSA-Net methodology was validated by conducting a series of classification accuracy tests using two publicly available eye disease datasets and was benchmark against a number of different pretrained convolutional neural network(CNN)architectures.The results showed that the ECSA-Net achieved classification accuracies of 60.00%and 69.92%,respectively,while using only a compact architecture with 0.56 million parameters.This represents a reduction in parameter size by a factor of 14×to 247×compared to other pretrained models.Additionally,the attention modules added to the architecture significantly increased sensitivity to disease-relevant regions of the retina while maintaining low computational cost,making ECSA-Net a viable option for real-time clinical use.ECSA-Net is both efficient and accurate in automating the classification of eye diseases,combining high performance with the ethical considerations of medical artificial intelligence(AI)deployment.The ECSA-Net frameworkmitigates algorithmic bias in training datasets and protects individuals’privacy and transparency in decision-making,thereby facilitating human-AI collaboration.The two areas of technical performance and ethical integration are needed for the responsible and scalable use of ECSA-Net in a variety of ophthalmic care settings.展开更多
基金partially funded by the Programa Nacional de Becas y Crédito Educativo of Peru and the Universitat de València,Spain.
文摘Self-Explaining Autonomous Systems(SEAS)have emerged as a strategic frontier within Artificial Intelligence(AI),responding to growing demands for transparency and interpretability in autonomous decisionmaking.This study presents a comprehensive bibliometric analysis of SEAS research published between 2020 and February 2025,drawing upon 1380 documents indexed in Scopus.The analysis applies co-citation mapping,keyword co-occurrence,and author collaboration networks using VOSviewer,MASHA,and Python to examine scientific production,intellectual structure,and global collaboration patterns.The results indicate a sustained annual growth rate of 41.38%,with an h-index of 57 and an average of 21.97 citations per document.A normalized citation rate was computed to address temporal bias,enabling balanced evaluation across publication cohorts.Thematic analysis reveals four consolidated research fronts:interpretability in machine learning,explainability in deep neural networks,transparency in generative models,and optimization strategies in autonomous control.Author co-citation analysis identifies four distinct research communities,and keyword evolution shows growing interdisciplinary links with medicine,cybersecurity,and industrial automation.The United States leads in scientific output and citation impact at the geographical level,while countries like India and China show high productivity with varied influence.However,international collaboration remains limited at 7.39%,reflecting a fragmented research landscape.As discussed in this study,SEAS research is expanding rapidly yet remains epistemologically dispersed,with uneven integration of ethical and human-centered perspectives.This work offers a structured and data-driven perspective on SEAS development,highlights key contributors and thematic trends,and outlines critical directions for advancing responsible and transparent autonomous systems.
文摘The integration of machine learning(ML)into geohazard assessment has successfully instigated a paradigm shift,leading to the production of models that possess a level of predictive accuracy previously considered unattainable.However,the black-box nature of these systems presents a significant barrier,hindering their operational adoption,regulatory approval,and full scientific validation.This paper provides a systematic review and synthesis of the emerging field of explainable artificial intelligence(XAI)as applied to geohazard science(GeoXAI),a domain that aims to resolve the long-standing trade-off between model performance and interpretability.A rigorous synthesis of 87 foundational studies is used to map the intellectual and methodological contours of this rapidly expanding field.The analysis reveals that current research efforts are concentrated predominantly on landslide and flood assessment.Methodologically,tree-based ensembles and deep learning models dominate the literature,with SHapley Additive exPlanations(SHAP)frequently adopted as the principal post-hoc explanation technique.More importantly,the review further documents how the role of XAI has shifted:rather than being used solely as a tool for interpreting models after training,it is increasingly integrated into the modeling cycle itself.Recent applications include its use in feature selection,adaptive sampling strategies,and model evaluation.The evidence also shows that GeoXAI extends beyond producing feature rankings.It reveals nonlinear thresholds and interaction effects that generate deeper mechanistic insights into hazard processes and mechanisms.Nevertheless,several key challenges remain unresolved within the field.These persistent issues are especially pronounced when considering the crucial necessity for interpretation stability,the demanding scholarly task of reliably distinguishing correlation from causation,and the development of appropriate methods for the treatment of complex spatio-temporal dynamics.
基金funded by the Deanship of Scientific Research(DSR)at King Abdulaziz University,Jeddah,under grant no.(GPIP:1074-612-2024).
文摘Log anomaly detection is essential for maintaining the reliability and security of large-scale networked systems.Most traditional techniques rely on log parsing in the reprocessing stage and utilize handcrafted features that limit their adaptability across various systems.In this study,we propose a hybrid model,BertGCN,that integrates BERT-based contextual embedding with Graph Convolutional Networks(GCNs)to identify anomalies in raw system logs,thereby eliminating the need for log parsing.TheBERT module captures semantic representations of log messages,while the GCN models the structural relationships among log entries through a text-based graph.This combination enables BertGCN to capture both the contextual and semantic characteristics of log data.BertGCN showed excellent performance on the HDFS and BGL datasets,demonstrating its effectiveness and resilience in detecting anomalies.Compared to multiple baselines,our proposed BertGCN showed improved precision,recall,and F1 scores.
文摘Over the past decade,the landscape of cybersecurity has been increasingly shaped by the growing sophistication and frequency of malware attacks.Traditional detection techniques,while still in use,often fall short when confronted with modern threats that use advanced evasion strategies.This systematic review critically examines recent developments in malware detection,with a particular emphasis on the role of artificial intelligence(AI)and machine learning(ML)in enhancing detection capabilities.Drawing on literature published between 2019 and 2025,this study reviews 105 peer-reviewed contributions from prominent digital libraries including IEEE Xplore,SpringerLink,ScienceDirect,and ACM Digital Library.In doing so,it explores the evolution of malware,evaluates detection methods,assesses the quality and limitations of widely used datasets,and identifies key challenges facing the field.Unlike existing surveys,this work offers a structured comparison of AI-driven frameworks and provides a detailed account of emerging techniques such as hybrid detection frameworks and image-based analysis.The findings indicate that AIbased models trained on diverse,high-quality datasets consistently outperform conventional methods,particularly when supported by feature engineering,explainable AI and a multi-faceted strategy.The review concludes by outlining future research directions,including the need for standardized datasets,enhanced adversarial robustness,and the integration of privacy-preserving mechanisms in malware detection systems.
文摘The increasing number of interconnected devices and the incorporation of smart technology into contemporary healthcare systems have significantly raised the attack surface of cyber threats.The early detection of threats is both necessary and complex,yet these interconnected healthcare settings generate enormous amounts of heterogeneous data.Traditional Intrusion Detection Systems(IDS),which are generally centralized and machine learning-based,often fail to address the rapidly changing nature of cyberattacks and are challenged by ethical concerns related to patient data privacy.Moreover,traditional AI-driven IDS usually face challenges in handling large-scale,heterogeneous healthcare data while ensuring data privacy and operational efficiency.To address these issues,emerging technologies such as Big Data Analytics(BDA)and Federated Learning(FL)provide a hybrid framework for scalable,adaptive intrusion detection in IoT-driven healthcare systems.Big data techniques enable processing large-scale,highdimensional healthcare data,and FL can be used to train a model in a decentralized manner without transferring raw data,thereby maintaining privacy between institutions.This research proposes a privacy-preserving Federated Learning–based model that efficiently detects cyber threats in connected healthcare systems while ensuring distributed big data processing,privacy,and compliance with ethical regulations.To strengthen the reliability of the reported findings,the resultswere validated using cross-dataset testing and 95%confidence intervals derived frombootstrap analysis,confirming consistent performance across heterogeneous healthcare data distributions.This solution takes a significant step toward securing next-generation healthcare infrastructure by combining scalability,privacy,adaptability,and earlydetection capabilities.The proposed global model achieves a test accuracy of 99.93%±0.03(95%CI)and amiss-rate of only 0.07%±0.02,representing state-of-the-art performance in privacy-preserving intrusion detection.The proposed FL-driven IDS framework offers an efficient,privacy-preserving,and scalable solution for securing next-generation healthcare infrastructures by combining adaptability,early detection,and ethical data management.
基金supported by the Institute of Information&Communications Technology Planning&Evaluation(IITP)grant funded by the Korea goverment(MSIT)(No.RS-2024-00439139,Development of a Cyber Crisis Response and Resilience Test Evaluation Systems)this research was supported by the MSIT(Ministry of Science and ICT),Korea,under the Graduate School of Virtual Convergence support program(IITP-2026-RS-2023-00254129)supervised by the IITP(Institute for Information&Communications Technology Planning&Evaluation)+1 种基金supported by the“Regional Innovation System&Education(RISE)”through the Seoul RISE Center,funded by the Ministry of Education(MOE)and the Seoul Metropolitan Government(2026-RISE-01-018-05)supported by QuadMiners Corp.
文摘Illicit web ecosystems,encompassing phishing,illegal online gambling,scam platforms,and malicious advertising,have rapidly expanded in scale and complexity,creating severe social,financial,and cybersecurity risks.Traditional rule-based and blacklist-driven detection approaches struggle to cope with polymorphic,multilingual,and adversarially manipulated threats,resulting in increasing demand for Artificial Intelligence(AI)-based solutions.This review provides a comprehensive synthesis of research on AI-driven threat detection for illicit web environments.It surveys detection models across multiple modalities,including text-based analysis of Uniform Resource Locator(URL)and HyperText Markup Language(HTML),vision-based recognition of webpage layouts and logos,graphbased modeling of domain and infrastructure relationships,and sequence modeling using transformer architectures.In addition,the paper examines system architectures,data collection and labeling pipelines,real-time detection frameworks,and widely used benchmark datasets,while also discussing their inherent limitations related to imbalance,representativeness,and reproducibility.The review highlights critical challenges such as evasion strategies,cross-lingual detection barriers,deployment latency,and explainability gaps.Furthermore,it identifies emerging research directions,including the use of Generative Adversarial Network(GAN)for threat simulation,few-shot and self-supervised learning for data-scarce environments,Explainable Artificial Intelligence(XAI)for transparency,and predictive AI for proactive threat forecasting.By integrating technical,legal,and societal perspectives,this survey offers a structured foundation for researchers and practitioners to design resilient,adaptive,and trustworthy AI-based defense systems against illicit web threats.
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.
文摘With the ongoing digitalization and intelligence of power systems,there is an increasing reliance on large-scale data-driven intelligent technologies for tasks such as scheduling optimization and load forecasting.Nevertheless,power data often contains sensitive information,making it a critical industry challenge to efficiently utilize this data while ensuring privacy.Traditional Federated Learning(FL)methods can mitigate data leakage by training models locally instead of transmitting raw data.Despite this,FL still has privacy concerns,especially gradient leakage,which might expose users’sensitive information.Therefore,integrating Differential Privacy(DP)techniques is essential for stronger privacy protection.Even so,the noise from DP may reduce the performance of federated learning models.To address this challenge,this paper presents an explainability-driven power data privacy federated learning framework.It incorporates DP technology and,based on model explainability,adaptively adjusts privacy budget allocation and model aggregation,thus balancing privacy protection and model performance.The key innovations of this paper are as follows:(1)We propose an explainability-driven power data privacy federated learning framework.(2)We detail a privacy budget allocation strategy:assigning budgets per training round by gradient effectiveness and at model granularity by layer importance.(3)We design a weighted aggregation strategy that considers the SHAP value and model accuracy for quality knowledge sharing.(4)Experiments show the proposed framework outperforms traditional methods in balancing privacy protection and model performance in power load forecasting tasks.
文摘Fingerprint classification is a biometric method for crime prevention.For the successful completion of various tasks,such as official attendance,banking transactions,andmembership requirements,fingerprint classification methods require improvement in terms of accuracy,speed,and the interpretability of non-linear demographic features.Researchers have introduced several CNN-based fingerprint classification models with improved accuracy,but these models often lack effective feature extractionmechanisms and complex multineural architectures.In addition,existing literature primarily focuses on gender classification rather than accurately,efficiently,and confidently classifying hands and fingers through the interpretability of prominent features.This research seeks to improve a compact,robust,explainable,and non-linear feature extraction-based CNN model for robust fingerprint pattern analysis and accurate yet efficient fingerprint classification.The proposed model(a)recognizes gender,hands,and fingers correctly through an advanced channel-wise attention-based feature extraction procedure,(b)accelerates the fingerprints identification process by applying an innovative fractional optimizer within a simple,but effective classification architecture,and(c)interprets prominent features through an explainable artificial intelligence technique.The encapsulated dependencies among distinct complex features are captured through a non-linear activation operation within a customized CNN model.The proposed fractionally optimized convolutional neural network(FOCNN)model demonstrates improved performance compared to some existing models,achieving high accuracies of 97.85%,99.10%,and 99.29%for finger,gender,and hand classification,respectively,utilizing the benchmark Sokoto Coventry Fingerprint Dataset.
文摘Feature selection(FS)plays a crucial role in medical imaging by reducing dimensionality,improving computational efficiency,and enhancing diagnostic accuracy.Traditional FS techniques,including filter,wrapper,and embedded methods,have been widely used but often struggle with high-dimensional and heterogeneous medical imaging data.Deep learning-based FS methods,particularly Convolutional Neural Networks(CNNs)and autoencoders,have demonstrated superior performance but lack interpretability.Hybrid approaches that combine classical and deep learning techniques have emerged as a promising solution,offering improved accuracy and explainability.Furthermore,integratingmulti-modal imaging data(e.g.,MagneticResonance Imaging(MRI),ComputedTomography(CT),Positron Emission Tomography(PET),and Ultrasound(US))poses additional challenges in FS,necessitating advanced feature fusion strategies.Multi-modal feature fusion combines information fromdifferent imagingmodalities to improve diagnostic accuracy.Recently,quantum computing has gained attention as a revolutionary approach for FS,providing the potential to handle high-dimensional medical data more efficiently.This systematic literature review comprehensively examines classical,Deep Learning(DL),hybrid,and quantum-based FS techniques inmedical imaging.Key outcomes include a structured taxonomy of FS methods,a critical evaluation of their performance across modalities,and identification of core challenges such as computational burden,interpretability,and ethical considerations.Future research directions—such as explainable AI(XAI),federated learning,and quantum-enhanced FS—are also emphasized to bridge the current gaps.This review provides actionable insights for developing scalable,interpretable,and clinically applicable FS methods in the evolving landscape of medical imaging.
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.
基金extend their appreciation to the Princess Nourah bint Abdulrahman University Researchers Supporting Project number(PNURSP2026R760)Princess Nourah bint Abdulrahman University,Riyadh,Saudi Arabia.The authors also extend their appreciation to the Deanship of Research and Graduate Studies at King Khalid University for funding this work through small group research under grant number RGP2/714/46.
文摘The convergence of Software Defined Networking(SDN)in Internet of Vehicles(IoV)enables a flexible,programmable,and globally visible network control architecture across Road Side Units(RSUs),cloud servers,and automobiles.While this integration enhances scalability and safety,it also raises sophisticated cyberthreats,particularly Distributed Denial of Service(DDoS)attacks.Traditional rule-based anomaly detection methods often struggle to detectmodern low-and-slowDDoS patterns,thereby leading to higher false positives.To this end,this study proposes an explainable hybrid framework to detect DDoS attacks in SDN-enabled IoV(SDN-IoV).The hybrid framework utilizes a Residual Network(ResNet)to capture spatial correlations and a Bi-Long Short-Term Memory(BiLSTM)to capture both forward and backward temporal dependencies in high-dimensional input patterns.To ensure transparency and trustworthiness,themodel integrates the Explainable AI(XAI)technique,i.e.,SHapley Additive exPlanations(SHAP).SHAP highlights the contribution of each feature during the decision-making process,facilitating security analysts to understand the rationale behind the attack classification decision.The SDN-IoV environment is created in Mininet-WiFi and SUMO,and the hybrid model is trained on the CICDDoS2019 security dataset.The simulation results reveal the efficacy of the proposed model in terms of standard performance metrics compared to similar baseline methods.
基金National Council for Scientific and Technological Development,Grant No.421278/2023-4,No.309248/2025-6。
文摘Coordinate transformation models often fail to account for nonlinear and spatially dependent distortions,leading to significant residual errors in geospatial applications.Here,we propose a residual-based neural correction(RBNC)strategy,in which a neural network learns to model only the systematic distortions left by an initial geometric transformation.By focusing solely on residual patterns,RBNC reduces model complexity and improves performance,particularly in scenarios with sparse or structured control point configurations.We evaluate the method using both simulated datasets(with varying distortion intensities and sampling strategies)and real-world image georeferencing tasks.Compared with direct neural network coordinate converters and classical transformation models,RBNC delivers more accurate and stable results under challenging conditions,while maintaining comparable performance in ideal cases.These findings demonstrate the effectiveness of residual modelling as a light-weight and robust alternative for improving coordinate transformation accuracy.
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Objective:Deep learning is employed increasingly in Gastroenterology(GI)endoscopy computer-aided diagnostics for polyp segmentation and multi-class disease detection.In the real world,implementation requires high accuracy,therapeutically relevant explanations,strong calibration,domain generalization,and efficiency.Current Convolutional Neural Network(CNN)and transformer models compromise border precision and global context,generate attention maps that fail to align with expert reasoning,deteriorate during cross-center changes,and exhibit inadequate calibration,hence diminishing clinical trust.Methods:HMA-DER is a hierarchical multi-attention architecture that uses dilation-enhanced residual blocks and an explainability-aware Cognitive Alignment Score(CAS)regularizer to directly align attribution maps with reasoning signals from experts.The framework has additions that make it more resilient and a way to test for accuracy,macro-averaged F1 score,Area Under the Receiver Operating Characteristic Curve(AUROC),calibration(Expected Calibration Error(ECE),Brier Score),explainability(CAS,insertion/deletion AUC),cross-dataset transfer,and throughput.Results:HMA-DER gets Dice Similarity Coefficient scores of 89.5%and 86.0%on Kvasir-SEG and CVC-ClinicDB,beating the strongest baseline by+1.9 and+1.7 points.It gets 86.4%and 85.3%macro-F1 and 94.0%and 93.4%AUROC on HyperKvasir and GastroVision,which is better than the baseline by+1.4/+1.6macro-F1 and+1.2/+1.1AUROC.Ablation study shows that hierarchical attention gives the highest(+3.0),followed by CAS regularization(+2–3),dilatation(+1.5–2.0),and residual connections(+2–3).Cross-dataset validation demonstrates competitive zero-shot transfer(e.g.,KS→CVC Dice 82.7%),whereas multi-dataset training diminishes the domain gap,yielding an 88.1%primary-metric average.HMA-DER’s mixed-precision inference can handle 155 pictures per second,which helps with calibration.Conclusion:HMA-DER strikes a compromise between accuracy,explainability,robustness,and efficiency for the use of reliable GI computer-aided diagnosis in real-world clinical settings.
基金supported and funded by the Deanship of Scientific Research at Imam Mohammad Ibn Saud Islamic University(IMSIU)under the grant number IMSIU-DDRSP2601.
文摘Ovarian cancer(OC)is one of the leading causes of death related to gynecological cancer,with the main difficulty of its early diagnosis and a heterogeneous nature of tumor biomarkers.Machine learning(ML)has the potential to process complex datasets and support decision-making in OC diagnosis.Nevertheless,traditional ML models tend to be biased,overfitting,noisy,and less generalized.Moreover,their black-box nature reduces interpretability and limits their practical clinical applicability.In this study,we introduce an explainable ensemble learning(EL)model,TreeX-Stack,based on a stacking architecture that employs tree-based learners such as Decision Tree(DT),Random Forest(RF),Gradient Boosting(GB),and Extreme Gradient Boosting(XGBoost)as base learners,and Logistic Regression(LR)as the meta-learner to enhance ovarian cancer(OC)diagnosis.Local Interpretable ModelAgnostic Explanations(LIME)are used to explain individual predictions,making the model outputs more clinically interpretable and applicable.The model is trained on the dataset that includes demographic information,blood test,general chemistry,and tumor markers.Extensive preprocessing includes handling missing data using iterative imputation with Bayesian Ridge and addressing multicollinearity by removing features with correlation coefficients above 0.7.Relevant features are then selected using the Boruta feature selection method.To obtain robust and unbiased performance estimates during hyperparameter tuning,nested cross-validation(CV)with grid search is employed,and all experiments are repeated five times to ensure statistical reliability.TreeX-Stack demonstrates excellent diagnostic performance,achieving an accuracy of 0.9027,a precision of 0.8673,a recall of 0.9391,and an F1-score of 0.9012.Feature-importance analyses using LIME and permutation importance highlight Human Epididymis Protein 4(HE4)as the most significant biomarker for OC.The combination of high predictive performance and interpretability makes TreeX-Stack a reliable tool for clinical decision support in OC diagnosis.
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.
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Princess Nourah bint Abdulrahman University Researchers Supporting Project number(PNURSP2026R77)Princess Nourah bint Abdulrahman University,Riyadh,Saudi Arabia,the Deanship of Scientific Research at Northern Border University,Arar,Saudi Arabia,through the project number NBU-FFR-2026-2248-01.
文摘Globally,diabetes and glaucoma account for a high number of people suffering from severe vision loss and blindness.To treat these vision disorders effectively,proper diagnosis must occur in a timely manner,and with conventional methods such as fundus photography,optical coherence tomography(OCT),and slit-lamp imaging,much depends on an expert’s interpretation of the images,making the systems very labor-intensive to operate.Moreover,clinical settings face difficulties with inter-observer variability and limited scalability with these diagnostic devices.To solve these problems,we have developed the Efficient Channel-Spatial Attention Network(ECSA-Net),a new deep learning-based methodology that integrates lightweight channel-and spatial-attention modules into a convolutional neural network.Ultimately,ECSA-Net improves the efficiency of computational resource use while enhancing discriminative feature extraction from retinal images.The ECSA-Net methodology was validated by conducting a series of classification accuracy tests using two publicly available eye disease datasets and was benchmark against a number of different pretrained convolutional neural network(CNN)architectures.The results showed that the ECSA-Net achieved classification accuracies of 60.00%and 69.92%,respectively,while using only a compact architecture with 0.56 million parameters.This represents a reduction in parameter size by a factor of 14×to 247×compared to other pretrained models.Additionally,the attention modules added to the architecture significantly increased sensitivity to disease-relevant regions of the retina while maintaining low computational cost,making ECSA-Net a viable option for real-time clinical use.ECSA-Net is both efficient and accurate in automating the classification of eye diseases,combining high performance with the ethical considerations of medical artificial intelligence(AI)deployment.The ECSA-Net frameworkmitigates algorithmic bias in training datasets and protects individuals’privacy and transparency in decision-making,thereby facilitating human-AI collaboration.The two areas of technical performance and ethical integration are needed for the responsible and scalable use of ECSA-Net in a variety of ophthalmic care settings.
