Air pollution poses a critical challenge to environmental sustainability, public health, and urban planning. Accurate air quality prediction is essential for devising effective management strategies and early warning systems. This study utilized a dataset comprising hourly measurements of pollutants such as PM2.5, NO_x, CO, and benzene, sourced from five metal oxide sensors and a certified analyzer in a polluted urban area, totaling 9,357 records collected over one year (March 2004–February 2005) from the Kaggle Air Quality Data Set. A comprehensive comparison of ten machine learning regression models XGBoost, LightGBM, Random Forest, Gradient Boosting, CatBoost, Support Vector Regression (SVR) with Bayesian Optimization, Decision Tree, K-Nearest Neighbors (KNN), Elastic Net, and Bayesian Ridge was conducted. Model performance was enhanced through Bayesian optimization and randomized cross-validation, with stacking employed to leverage the strengths of base models. Experimental results showed that hyperparameter optimization and ensemble strategies significantly improved accuracy, with the SVR model optimized via Bayesian optimization achieving the highest performance: an R² score of 99.94%, MAE of 0.0120, and MSE of 0.0005. These findings underscore the methodology’s efficacy in precisely capturing the spatial and temporal dynamics of air pollution.

Power transformers are critical components in ensuring the continuous and stable operation of power systems. Failures in these units can lead to significant power outages and costly downtime. Existing maintenance strategies often fail to accurately predict such failures, highlighting the need for novel predictive approaches. This study aims to improve the reliability of power systems by predicting transformer failures through the integration of IoT technologies and advanced machine learning techniques. The proposed hybrid model combines the LightGBM algorithm with GridSearch optimization to achieve both high predictive accuracy and computational efficiency. In addition, the model enhances interpretability by incorporating SHapley Additive exPlanations (SHAP) and Local Interpretable Model-Agnostic Explanations (LIME) for transparent decision making. The study presents a detailed comparison of different classification algorithms and evaluates their performance using metrics such as accuracy, recall, and F1 score. The results show that the hybrid model outperforms other methods, achieving an accuracy of 99.91%. The SHAP and LIME analyses provide engineers and researchers with valuable insights by highlighting the most influential features in failure prediction. In addition, the model’s ability to efficiently handle large data sets enhances its practicality in real-world power systems. By proposing an innovative approach to failure prediction, this research contributes to both the theoretical foundation and practical advancement of sustainable and reliable energy infrastructures.

This study introduces a hybrid deep learning model integrating EfficientNet-B3 and Vision Transformer with an Attention Fusion mechanism for automated colon cancer detection using the Kvasir endoscopic dataset. The model leverages EfficientNet-B3’s strength in capturing fine-grained local textures and Vision Transformer’s ability to model global contextual relationships. A multi-head attention-based fusion block harmonizes these features, achieving comprehensive representations and enhanced classification stability. Model optimization was guided by the Matthews Correlation Coefficient (MCC), alongside evaluations of accuracy, F1-score, and Brier Score. Experimental results demonstrate a 96.2% accuracy and an MCC of 0.961, surpassing standalone baselines and existing benchmark architectures. Cross-validation confirmed robust generalization, while Grad-CAM analyses improved interpretability by visualizing salient histopathological regions influencing predictions. Despite slight overfitting tendencies, the model maintained strong performance across all eight image classes. These findings highlight the model’s ability to address limitations of single-architecture approaches by combining local and global feature extraction, offering rapid, objective, and reliable diagnostic support. The proposed framework shows significant promise for integration into computer-aided colonoscopy systems, paving the way for enhanced clinical diagnostics and reduced pathologist workload through AI-driven precision medicine.

Accurate power prediction and fault detection in photovoltaic (PV) systems are essential for improving energy efficiency and enabling predictive maintenance. This study proposes a novel hybrid regression model based on a stacking ensemble architecture, which integrates multiple machine learning algorithms: histogram‐based gradient boosting (HGB), k‐nearest neighbors (k‐NN), decision tree (DT), random forest (RF), and LightGBM as base learners and employs Ridge regression as the meta‐learner. The model was designed to detect complex fault conditions such as partial shading and module‐level failures using SCADA‐type input features. The performance of the proposed model was evaluated using standard regression metrics ( R ² , RMSE, MAE), achieving superior results with an R ² of 0.9939, RMSE of 12.0184, and MAE of 8.0544. Paired t‐tests confirmed the statistical significance of performance improvements over baseline models ( p < 0.05). To ensure transparency, explainability analyses were conducted using SHapley Additive exPlanations (SHAP) and local interpretable model‐agnostic explanations (LIME), which revealed that fault‐related features had the greatest influence on model predictions. Comparative evaluation with recent state‐of‐the‐art approaches demonstrated that the proposed hybrid model is scalable, computationally efficient, and robust under varying environmental and operational conditions. The findings suggest that the model can serve as a reliable and interpretable solution for real‐time power forecasting and fault detection in PV systems.

Cervical cancer remains a major global health concern, highlighting the need for computer‐aided diagnostic systems that are both reliable and interpretable. Despite advances in deep learning–based cytology image classification, a gap persists in aligning model predictions with biologically meaningful explanations. This study aims to develop an explainability‐aligned, sample‐wise reliability‐weighted fuzzy ensemble framework for cervical cytology image classification to enhance both performance and interpretability. Methods: The proposed framework integrates three pretrained convolutional neural network backbones—InceptionV3, MobileNetV2, and Inception‐ResNetV2—within a fuzzy ensemble structure. A novel explainability metric, termed Explainable Artificial Intelligence Alignment (XAIHit), is introduced to quantitatively assess the spatial correspondence between Grad‐CAM activation maps and annotated cytoplasmic and nuclear regions. The model combines calibrated confidence estimates with XAIHit to produce a per‐sample reliability score that guides fuzzy aggregation, ensuring anatomically informed and statistically robust decision‐making. Experiments were conducted on the SIPaKMeD dataset. Results: The proposed ensemble achieved strong predictive performance, with accuracy ≈ 0.94, F1‐score ≈ 0.94, and area under the curve (AUC) ≈ 0.99. Calibration metrics further confirmed model reliability, with an expected calibration error (ECE) of 0.030, a Brier score of 0.078, and a negative log‐likelihood (NLL) of 0.198. The approach consistently outperformed conventional deep learning and fuzzy ensemble baselines. Conclusions: This study presents an interpretable and reliability‐aware fuzzy ensemble framework that advances AI‐assisted cervical cancer screening. By integrating explainability alignment and calibrated confidence into a unified reliability measure, the method fosters both diagnostic accuracy and clinical trust, marking a significant step toward safe, transparent medical AI systems. Comparable performance was also observed on an independent external validation dataset, confirming the cross‐dataset generalization capability of the proposed framework.