Using Artificial Intelligence Models to Assess Physical Activity for Children

This scientific study explores the application of machine learning algorithms in monitoring children's physical activity. The research analyzed the effectiveness of well-known supervised learning models such as Random Forest and Support Vector Machine. Based on data collected from wearable devices-such as accelerometers and gyroscope sensors-the physical activity levels of children were successfully classified into light, moderate, and high intensity. The results demonstrated that machine learning algorithms can serve as powerful tools for accurately identifying physical activity in children. This approach contributes to the development of a healthy lifestyle, continuous monitoring of physical conditions, and expands opportunities for practical implementation in the fields of education and healthcare. In particular, these technologies can be highly effective in assessing physical activity among children in schools and preschool institutions, as well as in the early identification of problems related to physical inactivity or obesity. The findings of the study provide a foundation for the future development of real-time, personalized monitoring systems and their integration with mobile applications or medical platforms. Moreover, this approach opens up possibilities for large-scale data analysis on children's movement patterns and its use in epidemiological research.