Publication Search

Recent Internet of Things (IoT) intrusion detection studies have reported near-perfect benchmark performance for Distributed Denial of Service (DDoS) detection, yet limited attention has been given to understanding how different traffic representations contribute to the detection process under highly imbalanced traffic conditions. This study presents an ablation-driven analysis to investigate the contribution of statistical and temporal representations for large-scale IoT DDoS detection using the CICIoT2023 dataset. Three experimental scenarios are evaluated, including statistical representation, temporal sequence representation, and hybrid statistical–temporal representation. Temporal representations are learned using a one-dimensional Convolutional Neural Network (1D-CNN) with lag-based traffic sequences, while ensemble tree-based classifiers are employed for final classification and representation analysis. In addition, multiple ablation configurations are designed to evaluate the impact of temporal dependency modeling and feature engineering strategies on detection performance. Experimental results show that statistical traffic representations remain highly effective for DDoS detection on CICIoT2023, achieving 99.36% accuracy and 99.31% weighted F1-score in the statistical representation scenario. Feature importance analysis further indicates that engineered statistical features contribute substantially more to the classification process than CNN-based temporal representations. Although temporal modeling captures sequential traffic behavior, its contribution is relatively limited and mainly acts as a complementary representation. Furthermore, the hybrid configuration produces only marginal improvements over the statistical representation alone. These findings highlight the importance of representation-level analysis for understanding the actual contribution of statistical and temporal modeling in modern IoT intrusion detection systems beyond relying solely on benchmark accuracy.

Tax administrations are undergoing a fundamental transition from conventional audit practices based on manual inspection and limited sampling toward data-driven supervision supported by big data analytics, artificial intelligence, and digital transaction infrastructures. However, developing economies, particularly in Southeast Asia, continue to face structural constraints such as fragmented legacy systems, informal economic activities, uneven digital literacy, corruption risks, weak data interoperability, and evolving privacy regulations. This study aims to develop a contextual framework for detecting potential tax-reporting fraud by integrating big data tax analytics, localized machine learning, explainable artificial intelligence, blockchain-enabled value-added tax data integrity, and socio-organizational governance. The study adopts a mixed-method sequential explanatory approach combined with Design Science Research. The methodological design integrates policy and institutional analysis, machine learning model design, and socio-organizational validation using secondary literature, Southeast Asian case studies, regulatory review, and simulated data architecture. The main contribution of this study is the Contextual Tax Analytics with AI and Blockchain Framework, or C-TAX-AIB Framework, consisting of three interrelated layers: Data Layer, Analytics Layer, and Governance and Human Layer. The Data Layer proposes a hybrid blockchain architecture for e-Faktur and value-added tax reporting integrity; the Analytics Layer introduces localized machine learning and explainable AI to support transparent risk scoring and anomaly detection; and the Governance and Human Layer embeds privacy protection, taxpayer digital literacy, auditor readiness, and trust-building mechanisms. The framework advances prior studies by moving beyond algorithmic fraud detection toward an integrated governance model suitable for developing economies. The study provides theoretical implications for public finance analytics and practical guidance for ASEAN tax administrations in designing accountable, explainable, and context-sensitive digital tax systems.

Artificial intelligence has the potential to support radiology workflows by assisting in the identification of cases that may require additional clinical attention. However, alert-oriented medical AI systems should provide not only classification outputs but also interpretable evidence that can be reviewed and audited by clinicians. This study develops and evaluates an explainable multimodal framework for binary chest X-ray alert classification using paired radiology reports and chest X-ray images. The text branch employs TF-IDF n-gram features with a class-balanced Logistic Regression classifier, while the image branch fine-tunes a pretrained ResNet18 model. The two branches are integrated through probability-level late fusion using a validation-selected fusion weight. Explainability is implemented in a modality-specific manner: global coefficient analysis is used to identify influential textual cues, while Grad-CAM heatmaps are used to visualize salient image regions. Experiments were conducted on paired samples from the Open-i/IU X-Ray dataset using text-only, image-only, and fusion-based evaluation settings. Additional analyses include case-level complementarity analysis, bootstrap confidence intervals for ROC-AUC, shortcut-feature inspection, and qualitative Grad-CAM auditing. The results indicate that the text modality provides the dominant predictive signal under the current proxy-label setting. Late fusion produced a small descriptive improvement on the test set, increasing accuracy from 0.8533 to 0.8667, F1-score from 0.8817 to 0.8936, and ROC-AUC from 0.8936 to 0.9025 compared with the text-only baseline. However, the observed ROC-AUC improvement was not statistically conclusive based on bootstrap analysis. These findings suggest that the proposed framework is useful as a reproducible and auditable multimodal prototype, while also highlighting important limitations, including proxy-label ambiguity, potential label leakage from radiology reports, limited image-branch contribution, lack of external validation, and the need for stronger explanation and calibration assessment.

Student dropout in higher education remains a persistent socioeconomic challenge, yet many predictive models reported in the literature are methodologically compromised by randomized cross-validation schemes that introduce temporal data leakage and artificially inflate predictive performance. This study proposes a longitudinal prescriptive learning analytics framework integrating three complementary methodological components: a Leave-One-Cohort-Out (LOCO) temporal validation protocol, a hybrid SMOTE-ENN class balancing strategy, and temporal velocity feature engineering derived from Learning Management System (LMS) behavioral trajectories. The framework was evaluated on a longitudinal dataset comprising 464,739 enrollment records and 77 features. Five predictive algorithms—XGBoost, LightGBM, CatBoost, Random Forest, and Logistic Regression—were comparatively assessed on a strictly isolated blind holdout cohort (2022), with CatBoost emerging as the champion estimator, achieving a PR-AUC of 0.8859, a Macro F1-Score of 0.9143, and the lowest Brier Score (0.0221), thereby demonstrating superior calibration and discriminative capability under severe class imbalance (93:7 ratio). Comprehensive ablation analysis revealed that temporal velocity features function not merely as additive predictors, but as a structural prerequisite enabling Synthetic Minority Oversampling Technique with Edited Nearest Neighbors (SMOTE-ENN) to generate high-quality synthetic boundary instances; removing these features reduced minority-class precision from 0.8302 to 0.6721. To operationalize predictive outputs into actionable intervention pathways, Diverse Counterfactual Explanations (DiCE) were implemented under a three-tier causal constraint architecture on 96 borderline high-risk students, generating 384 feasible intervention scenarios exclusively targeting forward-looking behavioral velocity metrics without constraint violations. Collectively, these findings advance the paradigm of prescriptive learning analytics by providing educational institutions with interpretable risk diagnostics and operationally feasible intervention guidance grounded in empirically validated behavioral and temporal dynamics.

Predictive maintenance (PdM) plays a crucial role in modern industrial systems by minimizing downtime, reducing maintenance costs, and optimizing asset performance. However, many predictive models operate as “black box” systems, limiting transparency and making it difficult for operators to interpret their outputs. This study aims to integrate Explainable Artificial Intelligence (XAI) techniques with Remaining Useful Life (RUL) prediction models to improve both accuracy and interpretability. Various machine learning and deep learning approaches, including Support Vector Machines (SVM), Random Forest (RF), XGBoost, Long Short-Term Memory (LSTM), and Convolutional Neural Networks (CNN), are employed to predict RUL using real-time sensor data from rotating machinery. XAI methods such as SHAP, LIME, and attention mechanisms are applied to provide human-understandable explanations of model predictions. The models are evaluated based on accuracy, Root Mean Square Error (RMSE), and interpretability scores. The results show that XAI-enhanced models outperform traditional approaches in predictive performance while offering greater transparency. These explanations help maintenance engineers better understand the factors influencing predictions, thereby improving decision-making and trust in the system. Nevertheless, the integration of XAI introduces additional computational complexity, which may pose challenges for large-scale industrial implementation. Overall, this study highlights the potential of combining XAI with RUL prediction to develop more reliable, transparent, and effective predictive maintenance solutions.

The retention of customers in the retail banking sector is a critical economic imperative; however, predictive modeling is frequently hindered by severe class imbalance and the “Black Box” nature of complex algorithms. This study proposes a Heterogeneous Stacking Ensemble framework integrating XGBoost, CatBoost, and Random Forest base learners with a Logistic Regression meta-learner to forecast customer attrition. To overcome the pervasive “Majority Class Bias,” we introduce a “Dual-Imbalance Defense” that synergizes the Synthetic Minority Over-sampling Technique (SMOTE) with algorithmic cost-sensitive penalization. Furthermore, moving beyond standard accuracy metrics, the framework mathematically derives a dynamic classification threshold to guarantee a strict 0.90 recall rate, actively optimizing the capture of at-risk capital. Model opacity is addressed through the integration of a SHapley Additive exPlanations (SHAP) TreeExplainer. This cooperative game theory approach provides localized, patient-level “Reason Codes” for regulatory compliance and reveals global systemic vulnerabilities, including non-linear drivers such as the “Product Paradox.” Achieving a 0.90 recall rate and an AUC of 0.8654, this framework provides a statistically robust and operationally transparent tool for targeted customer retention.

Manufacturing accounts for about 28.9% of global final energy use, making inefficient operations a major source of cost and greenhouse gas emissions. This review synthesizes how artificial intelligence supports energy optimization within Industry 4.0-enabled manufacturing systems. It organizes methods into four families: machine learning for forecasting and anomaly detection, deep learning for nonlinear and temporal modelling, reinforcement learning for adaptive scheduling and real-time control, and metaheuristics for balancing energy, throughput, and quality objectives. Applications span plant-level demand prediction and peak management, shop-floor rescheduling under dynamic pricing, equipment-level optimization through predictive maintenance, and system-wide planning using digital twins and cyber-physical integration. Reported benefits include lower energy costs, reduced downtime, improved productivity, and progress toward decarbonization. However, large-scale deployment is constrained by poor data quality and interoperability across IIoT, MES, ERP, and EMS platforms, high implementation and computational costs, skills gaps, and weak governance and benchmarking standards. Emerging solutions include federated learning and edge AI for privacy-preserving, low-latency analytics, explainable AI to enhance trust and auditability, tighter smart-grid integration, and circular economy-driven optimization. The review concludes with practical priorities for reliable, transparent, and scalable AI-enabled energy management.

Explainable artificial intelligence (XAI) has become a critical requirement in cybersecurity due to the high-stakes nature of security decision-making and the limitations of black-box learning models. This study investigates the construction of an explainable cybersecurity knowledge representation by leveraging standardized terminology from the NIST cybersecurity glossary. The primary problem addressed is the lack of transparent and semantically grounded reasoning mechanisms in existing AI-driven cybersecurity systems, which limits trust, accountability, and analyst adoption. To address this challenge, we propose a NIST-based semantic knowledge graph that embeds explainability directly into its ontology structure and reasoning process. The proposed framework systematically extracts definitional entities and relations from NIST glossary entries to construct a domain ontology and a multi-relational knowledge graph. A rule-based semantic relation extraction method is employed to ensure faithful, interpretable, and reproducible reasoning paths. The resulting knowledge graph contains over 3,000 cybersecurity concepts and approximately 27,000 semantic relations, covering hierarchical, associative, dependency, and mitigation semantics. Experimental evaluation demonstrates that the proposed approach achieves a high level of explainability, with 92.4% of reasoning outcomes being fully traceable and only 1.4% classified as non-traceable. Most explainable reasoning paths are limited to two or three hops, indicating an effective balance between inferential depth and human interpretability. Structural analysis further confirms the presence of meaningful hub concepts that support multi-hop semantic inference. These results confirm that ontology-driven, standard-based knowledge graphs provide a robust foundation for explainable cybersecurity intelligence. The study concludes that explainability-by-design, grounded in authoritative standards, offers a viable and trustworthy alternative to opaque AI models for cybersecurity applications.

This study examines the application of Transparent Artificial Intelligence (AI) for fraud detection in public welfare programs using publicly available administrative data. Persistent challenges in welfare governance such as misallocation, fraud, and data inaccuracy necessitate analytical frameworks that are both effective and explainable. The research aims to design and evaluate an interpretable anomaly detection system capable of identifying irregularities in welfare distribution while maintaining transparency and accountability. Methodologically, the study employs two unsupervised models Isolation Forest and Local Outlier Factor (LOF) to detect anomalies in sub-district-level welfare data, incorporating features such as population size, number of beneficiaries, and coverage ratio. An Explainable AI (XAI) framework integrating surrogate Random Forests, Permutation Feature Importance (PFI), and local linear surrogates (LIME-like) is applied to ensure interpretability of both global and local model behaviors. Findings reveal that receivers per 1000 population and percentage coverage are dominant determinants of anomaly scores. Fifteen administrative units were flagged for potential inconsistencies suggesting over- or under-reporting of beneficiaries. Cross-validation between IF and LOF models confirmed consistency in identifying anomalous regions. The integrated XAI explanations enhance transparency, enabling policymakers and auditors to trace the rationale behind detected anomalies. In conclusion, the proposed Transparent AI framework demonstrates that combining anomaly detection with interpretability tools can strengthen accountability and fairness in welfare administration. It offers a reproducible, ethical, and data-driven approach to social program monitoring, reinforcing public trust and supporting responsible AI governance.

Artificial Intelligence (AI) is increasingly recognized as a transformative enabler of early disease detection, with the potential to improve diagnostic accuracy, support predictive risk stratification, and advance preventive healthcare. Despite rapid methodological progress, many existing reviews remain performance-centric, offering limited insight into generalizability, ethical governance, and real-world implementation constraints. This paper presents a narrative and integrative review with an adoption-focused, translational perspective, synthesizing recent developments in AI-driven early disease detection across oncology, cardiology, neurology, and infectious disease surveillance. Drawing on peer-reviewed literature published primarily between 2016 and 2025, the review examines reported performance gains alongside persistent limitations related to data heterogeneity, population bias, explainability, and regulatory fragmentation. Through cross-sectional synthesis, we identify three recurring gaps in prior reviews: (i) overgeneralization of AI’s diagnostic superiority, (ii) insufficient consideration of ethical and legal accountability, and (iii) a lack of actionable guidance for scalable clinical implementation. Integrating technical, ethical, and policy dimensions into a unified conceptual framework, this review demonstrates that while AI systems can consistently enhance diagnostic accuracy and early risk stratification in well-defined tasks, sustained clinical adoption depends on aligning technical performance with governance readiness, interpretability, and workflow integration. The analysis further highlights how implementation mechanisms—such as explainable AI, continuous post-deployment monitoring, and clinician-centered deployment strategies—mediate the translation of algorithmic innovation into real-world healthcare impact. Overall, this review provides a critical reference for researchers, clinicians, and policymakers seeking to translate AI innovation into safe, equitable, and trustworthy clinical practice.

Pneumonia remains a leading cause of morbidity and mortality worldwide, particularly in resource-limited settings and among elderly populations, where timely diagnosis and continuous monitoring are often constrained by limited clinical infrastructure. This study presents an edge–cloud–integrated framework for early pneumonia risk monitoring, leveraging multimodal wearable sensors and deep learning to support continuous short-duration monitoring. The proposed system is designed to operate in near real time under simulated deployment conditions, continuously acquiring and analyzing physiological signals (respiratory rate, heart rate, SpO₂, and body temperature) alongside event-driven acoustic biomarkers (cough sounds) within a distributed architecture. A lightweight edge module performs local signal preprocessing and anomaly triage, selectively transmitting salient information to a cloud-based multimodal deep learning model for refined risk estimation and interpretability analysis. The framework was evaluated using a multi-source dataset comprising public repositories (MIMIC-III and Coswara) and a clinically supervised wearable study conducted in two Nigerian hospitals, resulting in 718  hours of quality-controlled multimodal monitoring data. In a pooled multi-source evaluation, the system achieved an AUC of 0.95, while in a clinically realistic local-only evaluation, the AUC was 0.86, reflecting a consistent but preliminary diagnostic signal. These results highlight the importance of local data adaptation for real-world applicability and suggest that multimodal AI can provide meaningful early risk indicators under resource constraints. Beyond predictive performance, this work demonstrates the feasibility of integrating multimodal learning, edge–cloud computation, and explainable analytics into a deployment-aware, privacy-preserving monitoring framework for low-resource healthcare environments.

Heart disease is the leading cause of death in Indonesia and globally, necessitating an early screening system that is both accurate and clinically trustworthy. Although XGBoost demonstrates high predictive performance, its black-box nature undermines clinical trust, while low recall risks missed diagnosis an unacceptable consequence in population screening, especially in middle-income countries with limited healthcare resources. This study aims to develop a sensitive, transparent, and implementation-ready heart disease screening framework through the integration of SHAP-based Explainable AI. The CDC's Indicators of Heart Disease dataset (319,795 samples) was processed according to WHO/CDC standards, followed by class imbalance handling, hyperparameter optimization using RandomizedSearchCV, evaluation based on metrics sensitive to minority classes (AUC, recall, F1-score, AUC-PR), and threshold tuning to maximize recall. The baseline model showed a very low recall of 12.18%. After optimization and threshold tuning at 0.10, the model achieved recall >96% (96.79%) with a G-mean of 0.7477, supported by SHAP interpretation stability and the ability to capture non-linear interactions between advanced age (AgeCategory_WHO) and poor general health (GenHealth). SHAP analysis confirmed the alignment of dominant features with medical evidence, and its visualizations provide transparent explanations for healthcare professionals indicating its potential implementation as an interpretable clinical decision support system.

This research explores the role of human algorithm interaction mechanisms in enhancing trust, reliability, and user confidence in Decision Support Systems (DSS). Traditional DSS models often focus solely on algorithmic accuracy and performance, neglecting crucial factors such as transparency and user engagement, which are essential for building trust. By incorporating explainable AI (XAI) techniques like SHAP and LIME, real-time feedback mechanisms, and user-friendly interfaces, the study develops structured interaction models that improve the interpretability of AI-driven decisions. The results show that transparent decision-making processes and interactive features significantly enhance user trust, making DSS more reliable and easier to adopt. Users interacting with systems that provide clear, understandable explanations of decisions, along with real-time updates on the system’s confidence, reported higher levels of decision-making confidence, especially in high-stakes scenarios. These improvements lead to greater user engagement and adoption of the system in various domains, including healthcare and finance. The study also highlights the importance of balancing interpretability with efficiency in user interface design to ensure both trust and usability. The findings contribute to the design of more user-centric DSS that prioritize trust, interpretability, and cognitive factors, providing a framework for the successful integration of intelligent decision support systems in complex decision-making environments. Future research should focus on refining interaction models and exploring the broader applicability of these systems in different sectors.

Autonomous decision-making systems increasingly rely on complex artificial intelligence models to operate in dynamic and safety-critical environments. While these models provide strong predictive capabilities, their black-box nature limits transparency, trust, and accountability. This study proposes a structured research methodology for integrating Explainable Artificial Intelligence (XAI) into autonomous decision-making systems. The research adopts a conceptual–analytical approach to develop an explainability-oriented framework that embeds transparency across perception, decision-making, and action execution stages. The methodology includes literature-driven problem identification, conceptual framework construction, classification and mapping of XAI methods, and formulation of explainability evaluation criteria. The results demonstrate that effective explainability in autonomous systems requires a hybrid integration strategy, combining in-model transparency with post-hoc explanation mechanisms. A structured mapping of XAI techniques to autonomous system components and a conceptual decision-flow diagram are presented to illustrate explainability integration. The findings highlight that layered and context-aware explainability enhances system interpretability, supports human oversight, and improves safety relevance without compromising autonomous operation. This study contributes a reusable methodological foundation for the design and evaluation of explainable autonomous systems, offering practical guidance for future empirical validation and real-world deployment in safety-critical applications.

Explainable Artificial Intelligence (XAI) has become a critical area of research within artificial intelligence, focusing on improving the transparency and interpretability of machine learning (ML) models, often referred to as "black-box" models. The need for XAI techniques arises from the inherent complexity of ML models, which can make their decision-making processes difficult for users to understand. This study investigates various XAI techniques, including LIME (Local Interpretable Model-agnostic Explanations) and SHAP (SHapley Additive exPlanations), to assess their impact on model interpretability without significantly compromising predictive performance. A comparative experimental design was used, applying these XAI methods to different ML models, including deep neural networks and ensemble methods, within large-scale enterprise data analytics systems. The results indicate that XAI methods significantly enhance model transparency and decision traceability, allowing users to understand the influence of individual features on predictions. While a slight reduction in predictive accuracy was observed, especially with simpler models, the trade-off between interpretability and performance was deemed acceptable, particularly in fields requiring transparency, such as healthcare, finance, and autonomous systems. The use of XAI in enterprise data systems has practical implications for fostering trust and enabling informed decision-making among stakeholders. Furthermore, the study discusses the challenges and limitations of applying XAI techniques, such as complexity, scalability, and model-specific limitations. Future research is suggested to focus on developing more scalable and efficient XAI methods, enhancing their applicability across various model types, and addressing the challenges of real-time applications. This will be crucial in ensuring the widespread adoption of XAI in critical domains, promoting the ethical use of AI while maintaining predictive accuracy.

  Abstract. This study aims to develop and evaluate an artificial intelligence–based expert system to support fault diagnosis in Air Handling Units (AHUs). Early fault identification in AHU systems is often constrained by reliance on technician experience, which may lead to inconsistent diagnostic outcomes. The proposed system is intended to provide a consistent, explainable, and practical decision-support tool to assist maintenance personnel, particularly less-experienced technicians, in identifying AHU faults accurately. Methodology: The research adopts a rule-based expert system approach using forward chaining inference. Knowledge acquisition was conducted through structured interviews with experienced HVAC technicians and supported by technical documentation. The resulting knowledge base consists of observable symptoms, diagnostic rules, and corresponding corrective actions. The system was implemented as an Android-based mobile application to enable direct field usage. System validation was performed using real operational fault scenarios, with expert diagnoses serving as the reference standard. Findings: Evaluation results indicate full agreement between the system-generated diagnoses and expert assessments across all tested scenarios. This demonstrates that the proposed system is capable of producing accurate and consistent diagnostic outcomes within its defined knowledge domain. Implications: The system operates using deterministic rules without incorporating uncertainty modeling or probabilistic reasoning. Additionally, validation was limited to a finite number of real-world scenarios, which may affect generalizability to broader AHU configurations. Practical implications: The expert system can be utilized as a practical diagnostic aid in routine AHU maintenance, improving response time, diagnostic consistency, and technician training effectiveness. Originality: This study contributes a mobile-based, explainable expert system specifically tailored for AHU fault diagnosis, emphasizing practical deployment and rule transparency rather than data-intensive learning models.

Background: The development of industrial technology in the Industry 4.0 era has encouraged the implementation of intelligent monitoring systems to improve machine reliability and operational efficiency. However, machine fault diagnosis systems based on artificial intelligence often face limitations in terms of interpretability because the models used are complex and difficult to explain. Objective: This study aims to develop a deep learning-based industrial machine fault diagnosis system integrated with an Explainable Artificial Intelligence (XAI) approach to improve diagnostic accuracy while providing interpretable insights for users. Method: The research method involves collecting data from industrial machine sensors consisting of vibration signals, temperature measurements, and acoustic signals, followed by data preprocessing and feature extraction processes. The processed data are then used to train a deep learning-based diagnostic model, after which explainability methods such as SHAP or LIME are applied to analyze the contribution of each feature to the model’s prediction results. Model performance is evaluated using accuracy, precision, recall, and F1-score metrics. Results: The results indicate that the proposed deep learning model achieves better performance compared to conventional machine learning methods such as Support Vector Machine and Random Forest. Furthermore, the explainability analysis reveals that vibration amplitude, increases in machine component temperature, and anomalies in acoustic signals are the main factors influencing machine fault detection. Therefore, the proposed system not only improves the accuracy of machine fault diagnosis but also provides transparency in the decision-making process, thereby supporting the implementation of predictive maintenance in smart manufacturing environments.

Hypertension is a major global health risk that requires accurate early detection, yet conventional methods struggle with complex and imbalanced health datasets. This study aims to optimize hypertension prediction using a Logistic Regression model integrated with Borderline-SMOTE to enhance recall and provide model transparency through SHAP (Shapley Additive Explanations). The method utilizes the BRFSS dataset, applying Borderline-SMOTE to address class imbalance at the decision boundary and XAI techniques for global and local interpretation. The findings show that the model achieved an accuracy of 0.719, an AUC of 0.800, and a significantly improved recall of 0.756. SHAP analysis identified age, high cholesterol, and BMI as the most influential risk factors, while waterfall plots successfully clarified individual risk extremes, ranging from 1.72% to 99.43% probability. These results imply that the proposed approach provides a sensitive and transparent screening tool for public health practitioners, effectively balancing statistical efficiency with clinical accountability.

Recommender systems play a critical role in shaping user decisions across digital platforms; however, the increasing complexity of recommendation algorithms has raised serious concerns regarding transparency, trust, and accountability. This study focuses on enhancing the transparency of recommender systems by integrating Explainable Artificial Intelligence (XAI) techniques within a MovieLens-based recommendation framework. The primary problem addressed is the opacity of conventional recommendation models, which limits user understanding of why certain items are recommended and may reduce trust, perceived fairness, and system acceptance. Accordingly, the main objective of this research is to design and evaluate a hybrid explainable recommender system that balances predictive accuracy with human-understandable explanations. The proposed approach combines Matrix Factorization, feature-importance-aware neural networks, and knowledge graph embeddings to construct a robust recommendation model. To enhance explainability, multiple XAI strategies are integrated, including model-agnostic methods (LIME, SHAP, and CLIME), argumentation-based explanations, and context-aware personalized explanations. A comprehensive evaluation framework is employed, incorporating algorithmic metrics (accuracy, fidelity, robustness, counterfactual consistency, and fairness) alongside human-centered evaluations measuring trust, transparency, cognitive load, and perceived usefulness. Experimental results demonstrate that the knowledge graph–enhanced hybrid model achieves superior recommendation accuracy compared to baseline approaches. Moreover, context-aware explanations consistently outperform other methods in terms of fidelity, robustness, and user-perceived transparency, while argumentation-based explanations are found to be the most persuasive. CLIME offers a strong balance between technical stability and interpretability. The findings indicate that no single explainability technique is universally optimal; instead, hybrid and adaptive explanation strategies are most effective. In conclusion, this study confirms that human-centered, context-adaptive XAI significantly improves transparency and user trust in recommender systems, highlighting explainability as a fundamental component rather than an optional enhancement.

End-to-end autonomous driving has emerged as a promising paradigm in which deep neural networks directly map raw visual inputs to continuous control actions. Despite its effectiveness, this approach suffers from limited transparency, posing significant challenges for deployment in safety-critical driving scenarios. This study addresses the lack of interpretability in vision-based end-to-end autonomous driving systems and aims to analyze model decision-making behavior under critical conditions such as sharp steering maneuvers and abrupt control transitions. To this end, an explainable end-to-end autonomous driving framework is proposed, combining a convolutional neural network trained via imitation learning with gradient-based visual attribution techniques, including Grad-CAM. The model predicts continuous steering, throttle, and braking commands directly from front-facing camera images, while explainability mechanisms are applied to reveal input regions influencing each control decision. Model performance is evaluated using both prediction accuracy and safety-oriented behavioral metrics. Experimental results show that the proposed explainable model achieves lower control prediction errors compared to a baseline end-to-end CNN, reducing steering mean squared error from 0.034 to 0.031, throttle error from 0.021 to 0.019, and brake error from 0.018 to 0.016. Moreover, safety-oriented analysis indicates improved driving stability, with steering variance reduced from 0.087 to 0.072 and abrupt control changes decreased from 14.6 to 10.3 events. Visual explanations consistently highlight road surfaces and lane-related structures during complex maneuvers, indicating reliance on semantically meaningful cues. In conclusion, the results demonstrate that integrating explainability into end-to-end autonomous driving not only preserves predictive performance but also correlates with smoother and more stable driving behavior. This framework contributes to the development of transparent and trustworthy autonomous driving systems suitable for safety-critical applications