In this work we have studied the criteria for assessing the quality of educational games and their impact on student learning outcomes by designing two experiments that were conducted in a duration of total of 6 years. The first experiment involves students in their final year of high school studies and the second on first year students in higher education. Our main results have shown that in comparison with traditional methods in education, teaching with the use of educational games increases the degree of mastering of educational material. It is also observed to increase the involvement and motivation of students to perform educational tasks by significantly reducing the time that will be required for reinforcement of the passed material. Additionally, it tends to cause less fatigue and holds their concentration much longer. Based on the results of our research, we have also identified the main criteria for evaluating the effectiveness of educational games. Our results suggest that a robust and comprehensive integration of educational games into the learning process will significantly improve the quality of education at all levels.
Glaucoma, cataracts, and diabetic retinopathy are among the world's major causes of blindness. Early detection and classification are the most important ways to prevent further loss of vision. With more and more medical imaging data and development in deep learning technologies, this is an excellent opportunity to automate this diagnostic process that significantly reduces the workload of healthcare professionals while at the same time increases the accuracy of diagnosis. The aim of this paper is to study the performance of a deep learning approach for the classification of eye diseases from retinal images. It is a CNN, specifically EfficientNet-B3, very effective in image classification tasks. The dataset used in this paper consists of retinal images that were divided into categories such as glaucoma, cataract, diabetic retinopathy, and normal ones. The research was performed by exploring the model performance in respect of accuracy, precision, recall, and F1-score in the classification of the data. The outcome showed that the model was able to achieve an overall accuracy of 92.5%, within which all metrics showed very outstanding performances in all categories. This discussion presents the effectiveness, shortcomings, and future possibilities the model may have in clinics for real-time diagnosis.
Dry Eye Disease (DED) is a widespread condition made worse by modern lifestyle factors such as extended screen time and reduced physical activity. This study focuses on developing a machine learning model, specifically a Random Forest classifier to predict the likelihood of DED based on user data. The dataset included features like screen time, eye discomfort, physical activity, weight, height, blood pressure and other symptoms.The model achieved an accuracy of 71.7\% and a log loss of 0.6501, showing moderatel performance in classifying DED. Feature importance shows that the most important features in prediction were eye strain, redness, irritation, screen time, and physical activity. These results align with clinical expectations and demonstrate that lifestyle-related data can be effectively used to detect DED risk.This research not only supports the idea that behavior affects eye health, but also shows how machine learning can be applied to assist in early detection and personalized recommendations for preventing DED.
Air pollution poses a significant health risk worldwide, with mortality rates from ambient particulate matter pollution increasing in many regions. This study focuses on forecasting air pollution-related mortality rates in two Central Asian cities, Bishkek (Kyrgyzstan) and Almaty (Kazakhstan). Utilizing time-series models, specifically Long Short-Term Memory (LSTM) networks and Prophet, the research aims to provide accurate predictions that can inform public health policies and interventions. The proposed methodology integrates advanced data preprocessing techniques, robust model architectures, and hyperparameter tuning to achieve an accuracy exceeding 85%. The findings reveal that time-series forecasting can effectively model the trend and seasonality of mortality rates, offering actionable insights for policymakers.
Many students in Kyrgyzstan, especially beginners, face difficulties accessing competitive programming platforms that are tailored to their needs. Existing systems primarily cater to advanced users or international audiences, making it harder for newcomers to engage effectively with algorithmic problem-solving. This project proposes the development of a competitive programming platform designed specifically for beginner learners. The system includes interactive problem statements, real-time code execution using external APIs, and automated evaluation of submitted solutions. The platform is built using a modern full-stack architecture with Angular, Spring Boot, and PostgreSQL, and integrates third-party services such as Judge0 for multi-language code execution. The goal is to encourage algorithmic thinking and programming skill development through a user-friendly and localized interface. Future enhancements will include a broader problem set, gamification elements, and educational resources such as tutorials and solution explanations.