Deep learning-based credit scoring systems face three interrelated challenges typically addressed in isolation: unavoidable class imbalance, model opacity, and demographic inequalities. This paper proposes the Explainable Fairness-Aware Deep Learning (EFADL) framework, which is a unified end-to-end pipeline designed to mitigate these risks simultaneously. The EFADL framework integrates three novel components: FC-SMOTE, a fairness-constrained oversampling module that preserves intra-group demographic balance; a multi-objective joint training loss combining focal imbalance correction with differentiable multi-attribute fairness penalties; and a dual-level SHAP module providing both instance-level adverse action explanations and group-level fairness attribution. Extensive experiments on the German Credit, Taiwan Credit, and LendingClub datasets demonstrate that EFADL achieves a better accuracy-fairness trade-off surface. Results indicate a 79% reduction in statistical parity and equal opportunity differences and a 70.5% decrease in maximum intersectional disparity, with a negligible AUC-ROC cost of only 1.2 percentage points. Furthermore, the framework reduces the fairness attribution gap by 71%, which provides evidence that it achieves fairness by suppressing reliance on demographic proxies rather than post-hoc calibration. By delivering stable, economically interpretable explanations, the EFADL framework aligns with the transparency requirements of the EU AI Act and US CFPB guidance and offers a deployable solution for regulatorily-compliant algorithmic lending.

EEG is an important source of information about the activity of the brain and finds broad applications in brain-computer interface and cognitive state analysis. This study focuses on the classification of eye states using EEG signals recorded from multiple scalp electrodes. Several machine learning methods were used to differentiate between eye open and eye closed conditions using EEG recordings. The models were tested using various classification measures to have a complete measure of predictive power. The K-Nearest Neighbors model is the highest performing in terms of classification with an accuracy of 0.9609 and ROC-AUC of 0.9927. CatBoost, XGBoost, and Random Forest were also found to be good predictors by ensemble methods. To enhance the transparency of the model, explainable artificial intelligence (XAI) models, including SHAP analysis, Partial Dependence Plot (PDP) and Individual Conditional Expectation Plot (ICE) plots were used to analyze features contribution and model performance. The results of the interpretability show that the signals of certain EEG electrodes, especially in frontal and occipital brain areas are important in the process of classifying the eye state. The results indicate that machine learning with explainable algorithms can be successfully used to assist EEG-based eye state detection in addition to providing insightful information on the decision-making of the model.

The process of navigating medical insurance markets, particularly within emerging economies characterized by financial volatility, has remained a significant hurdle for individual decision-makers. This study addresses the administrative complexity and cognitive burden inherent in selecting a medical insurance provider by transitioning the process from an intuitive task into a mathematically rigorous engineering decision solution. Utilizing a mixed-methods design, we developed a localized decision support system (DSS) in Python that integrates binary matchmaking logic with multi-criteria weighted scoring. This hybridized approach effectively handles the all-or-nothing nature of healthcare benefits that traditional models often struggle to resolve. Empirical justification for the system is anchored in a survey of 74 professionals within the case market, revealing a systemic mismatched budget paradox. The results revealed statistically significant negative correlations between income and healthcare risk drivers, such as age and family size, necessitating a recommendation engine capable of nuanced trade-off analysis. Validation results indicate that 80% of users experienced a marked improvement in decisional satisfaction and suitability. By prioritizing explainable AI over proprietary black-box systems, this study provides a scalable framework to mitigate informational asymmetry and financial risk, ensuring that healthcare access is dictated by personalized need rather than administrative complexity.

This paper presents a scientometric and analytical examination of machine learning applications in human factors research using a Scopus indexed dataset of 1,708 publications from 2019 to 2025. Keyword cooccurrence analysis was conducted using VOSviewer with a minimum occurrence threshold of 10, resulting in a network of 75 keywords grouped into eight thematic clusters. The thematic structure indicates a dominant empirical stream focused on cognitive state assessment, especially mental workload and fatigue detection using physiological signals such as electroencephalography and deep learning methods. Complementary clusters emphasize human AI interaction, explainable AI, decision support, automation and human reliability, affective computing, ergonomics in Industry 4.0/5.0 contexts, and healthcare oriented clinical decision support and patient safety. Overlay analysis shows recent growth in generative AI, large language models, fairness, and human centered AI concepts in 2024-2025. An analytical synthesis identifies major gaps, including limited human-in-the-loop validation, weak cross domain generalizability, insufficient longitudinal evaluation of human AI collaboration, and emerging explainability work that remains under validated empirically. The results provide a structured map of the field and a focused research agenda for responsible, transparent, and effective human centered intelligent systems.

The publication under consideration provides a detailed scientometric analysis of two hundred scientific publications, which were obtained from the Scopus database, and the main focus was the intersection of Magnetic Resonance Imaging (MRI), brain tumor research, and artificial intelligence (AI), or at least data science. The analytical process sharply defined the intellectual structure and development of the field from 2021 to 2025. Among the important observations was the predomination of the research interests directed at the deep learning models for segmentation and classification, the growing importance of Explainable AI (XAI) to mediate between the model's performance and clinical acceptance, and the flourishing of radiomics for oncological prediction. The geographical distribution of the field indicates that it is a world-wide activity with the major input coming from China, India, America, Germany, and Pakistan. The authors and institutions whose works have had the greatest impact are revealed, and they are frequently found being associated with the topics of taking the client in combination with data coming from different modalities through the process of data fusion and the clinical translation of technologies. This literature review maps out these trends in order to counsel the researchers who are just starting and in the meantime to indicate the directions of the future research, such as the need for solid and consistent; federated learning and creating lightweight models that are suitable for clinical deployment.