Enhancing action recognition in educational settings using AI-driven information systems for public health monitoring.

Introduction: The integration of Artificial Intelligence (AI) into educational environments is revolutionizing action recognition, offering a transformative opportunity to enhance public health monitoring. Traditional methods, which primarily rely on rule-based algorithms or handcrafted feature extraction, face significant challenges in adaptability, scalability, and real-time processing. These limitations hinder their effectiveness, particularly in detecting health-related behaviors such as sedentary patterns, social interactions, and hygiene compliance. Methods: To overcome these shortcomings, this research introduces an AI-driven information system that leverages advanced deep learning models and an Adaptive Knowledge Embedding Network (AKEN) to improve action recognition accuracy. The approach integrates AKEN with a Dynamic Personalized Learning Strategy (DPLS) to model student behaviors, predict future actions, and optimize intervention strategies by incorporating factors such as engagement levels, learning progress, and environmental conditions. Results: By utilizing reinforcement learning and explainable AI techniques, the system not only refines recognition accuracy but also ensures transparency in decision-making. Real-time engagement monitoring enhances adaptability, allowing educators and policymakers to make informed interventions. Discussion: Experimental results validate the system's superior performance over conventional approaches, demonstrating its ability to recognize complex behavioral patterns in educational settings. This innovation represents a significant step forward in AI-driven public health monitoring, fostering a safer and more responsive learning environment.