Research on Assessment and Evaluation Method Based on Sensor Technology and Machine Learning in College Physical Education Classroom

To further enhance the intelligence level of students' physical fitness assessment in university physical education classroom, and provide reliable basis for classroom emergency rescue, a physical fitness assessment model based on particle swarm optimization algorithm (PSO) improved BP neural network is proposed using sensor technology and machine learning method. Among them, taking BP neural network as the basic evaluation model architecture, the evaluation accuracy of the model is further improved by introducing PSO algorithm to optimize the training process and parameters of BP neural network. Simulation results reveal that the accuracy and feasibility of physical fitness assessment model based on PSO-BP neural network is higher, and its relative error is stable within 1% compared with the expert evaluation method. Compared to the evaluation model based on the traditional BP neural network, the evaluation accuracy of the constructed model has been significantly improved, which verifies the effectiveness of the optimization. In summary, the physical fitness assessment model based on PSO-BP neural network can be applied to actual physical education classroom teaching, thus accurate acquisition of student physical fitness levels is achieved, thereby providing reference for promoting student physical health, and providing reliable basis for emergency rescue related to the classroom.