Financial fraud in online credit card transactions poses significant challenges due to its increasing prevalence and the highly imbalanced nature of transactional data. This paper proposes a hybrid deep learning framework combining one-dimensional convolutional neural networks (CNN) and gated recurrent units (GRU) to effectively capture both spatial and temporal features of transaction sequences. Bayesian optimization is employed to fine-tune the model’s hyperparameters, improving detection performance without relying on synthetic oversampling. Evaluated on the widely-used European credit card fraud dataset, the proposed CNN-GRU model achieves superior results with an accuracy of 0.9996, an AUC-ROC of 0.9693, and an AUC-PR of 0.8709. These findings highlight the model’s robustness in identifying rare fraudulent transactions, outperforming several state-of-the-art methods and demonstrating the practical utility of deep learning combined with Bayesian optimization in fraud detection.

In the field of medical data analysis, the major challenges are high dimensionality and complexity due to temporal behavior of medical datasets. Feature selection is crucial to overcoming these challenges since it enhances interpretability, reduces processing expenses, and boosts model performance. To analyze the medical records, it is very essential to determine the most potential features that contribute more in classification or diagnosis of disease especially in the medical in the Medical Information Mart for Intensive Care III (MIMIC-III) dataset. This paper, highlights the importance of robust feature selection by developing a novel Hybrid feature selection framework that combines ReliefF and Ant Colony Optimization with Graph Neural Networks (ACO-GNN). The first step in the suggested approach is Ant Colony Optimization (ACO), which creates candidate feature subsets by effectively exploring the combinatorial feature space by mimicking pheromone-guided search behavior. Then, using graph-based representations of clinical variables, such as correlations between lab tests, drugs, and vital signs, Graph Neural Networks (GNNs) are used to model intricate, non-linear interactions among medical aspects. In order to ensure robustness and interpretability, ReliefF is used to rank and improve features by assessing their capacity to distinguish between patient outcome classes. The hybrid approach significantly outperforms conventional feature selection techniques like K-Nearest Neighbors with ReliefF (KNN-ReliefF) and XGBoost with SHAP Feature Importance (XGB-SHAP) in through tests on the MIMIC-III dataset, predictive performance indicators such as precision, accuracy, F1-score, recall, and AUC-ROC. The selected feature subsets offer clinically meaningful insights into critical factors influencing patient outcomes in intensive care, underscoring the potential of the ACO-GNN-ReliefF method for advancing predictive analytics and clinical decision support systems in healthcare.

The usage of social media to interchange ideas and facts has increased exponentially due to technological advancements. Platforms for video sharing, like YouTube, have distinctive environments and architecture that people use for entertainment, education, and to keep themselves updated. YouTube is one of the most frequently used social media platforms, and users can connect to it by viewing, sharing opinions through comments, liking and disliking videos. A viewpoint or judgement formed about anything is referred to as an opinion. It can be collected and used to check knowledge, suggest the author with new video ideas, and analyze user behaviour. In this study, the data extracted from the free video-sharing platform YouTube concerning the ‘Air India Flight Urination Case’ was observed recently to recognize people’s opinions on national and international levels. Based on approximately 10,000 comments about the incident, models are applied to classify and investigate the sentiments. This investigation uses TF-IDF and Bag of Words (BoW) text modelling techniques and observed that BoW performs better than TF-IDF. Moreover, Naive Bayes, Logistic Regression, Decision Tree, Support Vector Machines, and some ensemble algorithms like Random Forests, Gradient Boost, and Voting Classifier combining (Support Vector Machine, Decision Tree, Logistic Regression and Random Forest) with soft and hard voting had been applied and found that Support Vector Machine has the highest classification accuracy of 84%.

Rapid urbanization and industrialization have significantly increased air pollution levels, adversely affecting public health and environmental sustainability. Traditional monitoring systems often suffer from limited spatial coverage and delayed data analysis, making real-time pollution assessment challenging. Existing approaches struggle to efficiently process the large-scale, heterogeneous data generated by IoT-enabled Cognitive Radio-Based Sensor Networks (CRSNs). Conventional machine learning models often fail to capture complex temporal and spatial patterns in pollution dynamics, limiting predictive accuracy and early warning capabilities. This study proposes a transformer-based deep learning framework combined with IoT- enabled CRSNs for accurate and real-time air pollution monitoring. The CRSN comprises distributed sensors collecting continuous data on particulate matter (PM2.5, PM10), NO2, CO, O3, and other pollutants. The transformer model controls its self-attention mechanism to capture temporal dependencies and inter-sensor correlations, enabling robust prediction of pollution trends. Data preprocessing involves normalization, anomaly detection, and feature embedding to enhance model performance. Comparative experiments are conducted against conventional LSTM and GRU models to evaluate prediction accuracy and system responsiveness. Experimental results establish that the transformer-based model achieves superior performance with a mean absolute error (MAE) of 4.2 µg/m³ for PM2.5 prediction, outperforming LSTM (MAE = 6.1 µg/m³) and GRU (MAE = 5.8 µg/m³). The proposed system provides accurate, fine- grained pollution maps in real-time, enabling timely alerts and informed decision-making for environmental authorities.

Rapid urbanization has intensified air pollution and climate-related risks, challenging sustainable city planning. Green infrastructure (GI) has emerged as a vital strategy for mitigating these environmental stressors. However, selecting optimal GI interventions requires the integration of multiple, often conflicting, criteria such as pollution reduction efficiency, cost-effectiveness, and resilience to climate variability. Conventional decision-making approaches in urban planning often struggle to handle the uncertainty and complexity inherent in multi-criteria environmental assessments. This limitation hampers the identification of effective GI solutions tailored to specific urban contexts, leading to suboptimal pollution and climate risk mitigation. This study proposes a Deep Neural Fuzzy Multi-Criteria Decision Support System (DNF-MCDSS) for prioritizing GI strategies in urban environments. The framework combines fuzzy logic with deep neural networks to model uncertainty and non-linear relationships among environmental, social, and economic criteria. Input data covering air quality indices, climatic variables, land use patterns, and socioeconomic factors are processed through the hybrid network to generate a ranked list of GI interventions. The model’s performance is evaluated using case studies from metropolitan regions, with validation against expert assessments and conventional multi-criteria decision methods. Experimental results demonstrate that the proposed DNF-MCDSS consistently identifies GI strategies that maximize pollution reduction while enhancing climate resilience. For example, green roofs, urban tree corridors, and constructed wetlands were prioritized in scenarios with high pollution loads and extreme heat events. Compared with traditional weighted-sum and AHP methods, the framework achieved a 15–20% improvement in alignment with expert recommendations, showing its ability to capture complex interdependencies and uncertainty in urban environmental planning.

One of the most challenging issues these days is managing massive amounts of data that must be examined. In data mining applications, feature selection is extremely crucial. Feature Selection picks the fewest characteristics from many features requiring more calculation time, vast space, etc. Feature selection has captivated the interest of many researchers working on machine learning and data mining since it allows classifiers to be faster, more cost-effective, and more accurate. The previous study proposes a particle swarm optimization (PSO) approach with a few drawbacks. It can simply move into a local optimum and include minimum convergence ratio. However, when used to handle high-dimensional and complex problems, PSO’s computational complexity is acceptable. To address the issue to choose a subsection of characteristics with minimal redundancy &maximal relevance to classification, this study proposes a hybrid IPSO with a K-means technique. Initially, to normalize data, Z-score method is used. To enhance the accuracy of classifier, hybrid attribute extraction strategy is designed. Finally, the Support Vector Machine-based classifier is used to rank feature selection approaches based on their classification accuracy for a specific dataset. In this case, two datasets are used: WDBC and Hepatitis. According to the simulation findings, the suggested approach yields better efficiency than the traditional technologies.

Multi-Agent Reinforcement Learning (MARL) has emerged as a key paradigm for solving complex real-world problems involving multiple agents interacting in dynamic environments. However, training MARL models, especially for cooperative reasoning tasks, remains computationally intensive and sample-inefficient due to nonstationarity, credit assignment, and policy coupling issues. Conventional policy gradient methods struggle with convergence and scalability in multi-agent settings. Centralized training frameworks suffer from bottlenecks and synchronization overheads. Evolutionary algorithms, while more robust to non-differentiable objectives, are often too slow when applied in single-node environments. To address these challenges, we propose Distributed Co-evolutionary Policy Optimization (DCPO), a hybrid learning framework that distributes evolutionary computation across multiple nodes. DCPO decomposes the global policy search into sub-population-based parallel explorations, with each node evolving a subset of agent policies using fitness-driven mutation, crossover, and local policy gradient updates. A global coordinator aggregates top-performing policies periodically to ensure cooperative learning convergence. DCPO was tested on standard cooperative MARL benchmarks such as StarCraft II Micromanagement and Multi-Agent Particle Environments (MPE). Compared to traditional baselines such as MADDPG, QMIX, MAPPO, COMA, and EPOpt, DCPO showd up to 37% faster convergence, 25% higher final cumulative rewards, and enhanced generalization in unseen environments.

Autism spectrum disorder (ASD) encompasses a wide range of complex neurodevelopment conditions, posing significant challenges to early diagnosis. In fact, other than early diagnosis (i.e., before normal development) it is extremely challenging to adequately classify these subjects as there are no behavioral markers that are unique to ASD vs. non-ASD subjects. But the early detection is very important in helping with some of the developmental trajectories as the timely intervention which they are getting and the access which the kids and families have to specially designed therapies and support systems will be better. Neuroimaging in autism: a comprehensive review from a machine learning perspective. Autism Research- We investigate the effectiveness of CNN models and a hybrid CNN-RNN architecture in classifying ASD based on the neuroimaging data with an aim of improving feature extraction from the neuroimaging data using the Automated Anatomical Labeling (AAL) map. A key challenge in this task is how to build representational feature maps that can elucidate the intricate patterns across brain regions linked with ASD. Various kinds of models are tested VGG16 2D CNN, VGG16 3D CNN, EfficientNetV2, Inception v3, ResNet50 2D CNN and Hybrid model (CNN-RNN). While going for traditional CNN models gave at least 60% to 69% accuracies, the accuracy from hybrid CNN-RNN model, which combines spatial and temporal features, far outperformed other models with an accuracy of 100%. Thus, hybrid deep learning architectures have a substantial role in the classification of ASD. An involvement of AAL mapping and deep learning methods in this study proposed that the semi-automated approach could be implemented in forensic precision for the early diagnosis of ASD. These findings guide future neuroimaging-based diagnostic research and support the development of more sophisticated hybrid models for classifying autism spectrum disorder.

Real-world optimization problems in engineering, logistics, and resource allocation are often constrained and multi-modal, posing a challenge for traditional optimization algorithms. Metaheuristic algorithms inspired by natural and artificial phenomena have shown promise, but many fail to balance exploration and exploitation effectively, especially under stringent constraints. Existing algorithms such as Genetic Algorithms (GA), Particle Swarm Optimization (PSO), and Differential Evolution (DE) face issues in convergence speed and constraint handling, particularly in high-dimensional spaces or when constraints are dynamic or complex. We propose an Adaptive Pattern-Driven Optimization (APDO) algorithm, a novel tailor-inspired metaheuristic that mimics the adaptive decision-making process of a tailor designing garments. APDO integrates three primary operators—Pattern Selection, Fabric Adjustment, and Stitch Reinforcement—to handle constraints adaptively. The algorithm combines pattern memory (historical bests), probabilistic pattern mutation, and a constraint-domination principle to ensure feasibility and diversity. The core idea is to iteratively “cut and stitch” solutions to adapt the search process, enabling dynamic constraint satisfaction and global optimization. We benchmarked APDO against five popular methods (GA, PSO, DE, Firefly Algorithm, and Whale Optimization Algorithm) on a suite of 10 real-world constrained problems, including mechanical component design and energy scheduling tasks. APDO outperformed all baselines in terms of convergence speed, constraint satisfaction rate, and solution quality. In particular, APDO achieved an average feasibility rate of 97.6% and an improvement of 4.2–11.8% in best fitness across problems.

Global warming, pollution, and poor air quality are some of the main environmental problems. Many people worldwide, including miners, oil field workers, sailors, and industrial workers, operate in high-stakes situations where it is essential to keep an eye on their surroundings. In order to prevent hazards or disasters, it becomes essential to routinely check meteorological parameters such air quality, rainfall, water level, pH value, wind direction and speed, temperature, atmospheric pressure, humidity, soil moisture, light intensity, and turbidity. Appropriate monitoring is required for maintaining a healthy society and achieve sustainable growth. The usage of Internet of Things (IoT) which has been connected in recent years to the development of new sensors and environment monitoring systems (EMS) encompasses many devices for monitoring results in real-time information about environmental factors like Air temperatures. IoT devices may gather a wide range of data points for environmental monitoring, including temperature, radiation levels, and contaminant levels, giving a complete picture of the planet’s condition. One of the most important factors for safe and efficient operations in pipelines, refineries, and oil fields is temperature forecasting. Monitoring systems employ sensors to understand environmental temperature variations and mitigating the impacts of climate changes including global warming must be of utmost importance with commitment to address its dreadful implications to ensure global sustenance. IoT usage which has had significant impact on raising environmental standards can be used to track weather changes. In this context, the goal of the current paper is to conduct a critical evaluation of significant contributions and research projects on deep learning, IoT data, and EMS. An EMS based on deep learning (DL) is proposed in this study; the proposed system is complex, accurate, efficient, economical, and reliable. In order to ensure worker safety throughout implementation, our effort focuses on making precise temperature projections using past data. Temperature Assessments using Deep Learning Technique (TADLT), the suggested schema, makes predictions with an accuracy of above 90% by utilizing deep learning principles.