A Multivariate Evaluation Model of Physical Education Teaching Quality with Random Matrix Optimization Neural Network

Abstract

The status of physical education courses in education is increasing, there are many problems in teaching traditional physical education courses, and information technology and information-based education are developing continuously, providing enough space for the research of the topic. The online teaching of public physical education courses in colleges and universities is very complex, and its quality evaluation faces many difficulties. Considering the problem that most traditional PE teaching quality evaluation methods are affected by the accuracy of the established model in the process of positioning, a PE teaching quality evaluation model based on random matrix theory is established by using the advantage that random matrix theory does not rely on the simplification and assumptions of the system model and can extract effective information from a large amount of data. Firstly, the influencing factor data and physical education state data are constructed as a state augmentation matrix. Then, the average spectral radius of the characteristic statistic is used to construct indicators to analyze the size of the correlation between each influencing factor and physical education state to achieve the fundamental positioning of physical education quality. The main influencing factors of online teaching of public physical education classes in colleges and universities mainly include teacher quality, teaching process, course resources, and course effect, and the construction of an evaluation index system should be implemented based on adhering to the scientificity, accessibility, independence, and generality from these four aspects to construct the evaluation index system and evaluation criteria of online teaching quality of public physical education class in colleges and universities, which provides the final realization of the goal of physical education class in colleges and universities.