Prior studies have mostly studied artificial intelligence AI education from students’ perspectives from various majors. However, it is necessary to understand physical education students’ perception and their attitudes towards AI because they are essential personnel for physical education. To fill the gap in literature, the aim of this study is to determine physical education students’ attitudes and to analyze the perceived usefulness of AI. The study also investigated the differences in students’ attitudes and their perceived usefulness level regarding AI technology. A descriptive research design was used in this study, with a questionnaire to collect data from the study participants. The number of participants was 210 physical education students. The findings revealed that the participants expressed low to moderate positive attitudes towards using AI in their learning. Additionally, students had a moderate level of perceived usefulness of using AI. In addition, student attitudes and their perceived usefulness level regarding AI technology did not differ significantly based on gender, students age and their experiences. These findings provide valuable insight for academicians by filling a knowledge gap in the existing literature and demonstrating the concrete use of AI technology in the realm of physical education. Recommendations are enumerated based on the obtained findings.

Diabetic retinopathy (DR) is the main ocular complication of diabetes. Asymptomatic for a long time, it is subject to annual screening using dilated fundus or retinal photography to look for early signs. Fundus photography and optical coherence tomography (OCT) are used by ophthalmologists to assess retinal thickness and structure, as well as detect edema, hemorrhage, and scarring. The effectiveness of ConvNet no longer needs to be demonstrated, and its use in the field of imaging has made it possible to overcome many barriers, which were until now insurmountable with old methods. Throughout this study, a robust and optimal deep ConvNet is proposed to analyze fundus images and automatically distinguish between healthy, moderate, and severe DR. The proposed model combines the use of the ConvNet architecture taken from ImageNet, data augmentation, class balancing, and transfer learning in order to establish a benchmarking test. A significant improvement at the level of middle class which corresponds to the early stage of DR, which was the major problem in previous studies. By eliminating the need for retina specialists and broadening access to retinal care, the proposed model is substantially more robust in objectively early staging and detecting DR.

Human activity recognition (HAR) is the task of classifying sequences of data into defined movements. Taking advantage of deep learning (DL) methods, this research investigates and optimizes neural network architectures to effectively classify physical activities from smartphone accelerometer data. Unlike traditional machine learning (ML) methods employing manually crafted features, our approach employs automated feature learning with three deep learning architectures: Convolutional Neural Networks (CNN), CNN-based autoencoders, and Long Short-Term Memory Recurrent Neural Networks (LSTM RNN). The contribution of this work is primarily in optimizing LSTM RNN to leverage the most out of temporal relationships between sensor data, significantly improving classification accuracy. Experimental outcomes for the WISDM dataset show that the proposed LSTM RNN model achieves 96.1% accuracy, outperforming CNN-based approaches and current ML-based methods. Compared to current works, our optimized frameworks achieve up to 6.4% higher classification performance, which means that they are more appropriate for real-time HAR.

Understanding customer attitudes has become a critical component of decision-making due to the growing influence of social media and e-commerce. Text-based opinions are the most structured, hence playing an important role in sentiment analysis. Most of the existing methods, which include lexicon-based approaches and traditional machine learning techniques, are insufficient for handling contextual nuances and scalability. While the latter has limitations in model performance and generalization, deep learning (DL) has achieved improvement, especially on semantic relationship capturing with recurrent neural networks (RNNs) and convolutional neural networks (CNNs). The aim of the study is to enhance opinion mining by introducing a hybrid deep neural network model that combines a bidirectional gated recurrent unit (BGRU) and long short-term memory (LSTM) layers to improve sentiment analysis, particularly addressing challenges such as contextual nuance, scalability, and class imbalance. To substantiate the efficacy of the proposed model, we conducted comprehensive experiments utilizing benchmark datasets, encompassing IMDB movie critiques and Amazon product evaluations. The introduced hybrid BGRU-LSTM (HBGRU-LSTM) architecture attained a testing accuracy of 95%, exceeding the performance of traditional DL frameworks such as LSTM (93.06%), CNN+LSTM (93.31%), and GRU+LSTM (92.20%). Moreover, our model exhibited a noteworthy enhancement in recall for negative sentiments, escalating from 86% (unbalanced dataset) to 96% (balanced dataset), thereby ensuring a more equitable and just sentiment classification. Furthermore, the model diminished misclassification loss from 20.24% for unbalanced to 13.3% for balanced dataset, signifying enhanced generalization and resilience.

One major factor influencing the development of eco-friendly policies and the implementation of climate change mitigation strategies is the accurate projection of CO₂ emissions. Traditional statistical models face significant limitations in capturing complex nonlinear interactions within high-dimensional emissions data. Advanced deep learning architectures offer new opportunities to overcome these computational challenges due to their strong pattern-recognition capabilities. This paper evaluates four distinct deep learning architectures for CO₂ emissions forecasting: Long Short-Term Memory (LSTM) networks, Gated Recurrent Units (GRUs), Hybrid Convolutional–LSTM (CNN–LSTM) systems, and Dense Neural Networks (DNNs). A comprehensive comparison is conducted using consistent training protocols, hyperparameters, and performance metrics across five prediction horizons (1, 3, 6, 12, and 24 steps ahead) to reveal architecture-specific degradation patterns. Furthermore, analyzing emissions by category provides insight into the suitability of each architecture for varying levels of pattern complexity. LSTM-based models demonstrate particular strength in modeling long-term temporal dependencies, making them well-suited for integration into long-range environmental policy planning frameworks. Overall, this study provides empirical evidence supporting the use of neural networks in climate modeling and proposes criteria for selecting optimal architectures based on forecasting horizon and computational constraints.

Employee retention is still among the biggest issues in contemporary human resource management with the average cost of staff turnover ranging from 20 to 200 % of the yearly salary depending on the job complexity. This article provides a comprehensive machine learning framework for the organizational data-driven prediction of employee satisfaction and turnover comprising 6,000 employees from different departments. We built two interdependent predictive models: (1) a regression model for predicting job satisfaction, and (2) a classification model to predict whether employees will leave their jobs or stay. The regression model attained a SMAPE (symmetric mean absolute percentage error) of 13.40 % using random forest with optimized hyperparameters, thereby successfully fulfilling the rigorous target criterion of ≤15 %. The classification model, implemented using gradient boosting and feature selection, achieved an ROC-AUC score of 0.9314, significantly exceeding the target threshold of 0.91. Thorough feature importance analysis shows that the supervisor evaluation is the variable most highly correlated with job satisfaction, while junior-level employees have much higher attrition rates. The innovative synthetic job satisfaction feature, based on the optimized regression model, was the leading source of influence in the classification of attrition. The robustness and ability to generalize can also be ensured by cross-validation, which was examined by using statistical testing. The findings support smart workforce management from a human-centric, data-driven, and quantitative perspective, acting as substantial evidence that may communicate the feasibility of using ML in smart workforce management applications.