Optimizing College Physical Education Theory Instruction with The Help of Big Data

Abstract

In order to improve the teaching effect of physical education (PE)theory courses in colleges and universities (CAU), this paper combines big data (BD)technology to construct a teaching model (TM)system for PE theory courses in CAU. In this paper, the sensor information is filtered to remove noise and outliers, and the initial simple fusion of sensor information is realized. Moreover, this paper fuzzies the partitioned data, which effectively avoids the problem of too large sample space caused by the direct input of ultrasonic data into the neural network (NN). In addition, this paper analyzes the sensor information fusion process based on BP NN (BPNN), and constructs the sample space by encoding the data. Finally, this paper uses the L-M optimization algorithm instead of the gradient descent method to train the NN to improve the iterative efficiency. The experimental research shows that the teaching mode system of PE theory course proposed in this paper can effectively improve the teaching effect of PE theory course.