Evaluation model of physical education teaching effect based on machine learning algorithm with biomechanical integration

This study aims to propose a machine learning-based approach to assess the impact of physical education instruction, integrating biomechanical principles to addressing the challenge of difficult and inconsistent result evaluation. The evaluation index system is constructed based on key factors influencing teaching quality, including teaching attitude, content delivery, instructional methods, and teaching effects, with an added emphasis on biomechanical metrics such as movement efficiency, joint kinematics, and muscle activation patterns. This paper suggests a Biomechanically Enhanced Physical Education based Dwarf Mongoose Optimization Algorithm (PE-DMOA) for the advancement of PE, leveraging ML technology and biomechanical analysis to optimize PE teaching strategies. We pre-process the physical education dataset, incorporating biomechanical data from motion capture systems, force plates, and electromyography (EMG), alongside traditional teaching metrics gathered information from 2150 students and 72 teachers across various schools, to predict learning efficiency more accurately than previous methods. With a student satisfaction rate of 95.6%, the experimental results confirm the efficiency of the evaluation model developed in this article. The study’s results show that the suggested model (PE-DMOA) is 98.5% accurate. This means that it helps to look into the effects of machine learning and biomechanics on physical education teaching and gives good recall, accuracy, and precision results. Educators and learners can utilize the PE-DMOA evaluation model to enhance the quality and efficiency of instruction, streamline administrative tasks, and advance the scientific, standardized, and specialized delivery of physical education in classrooms.