Internal teaching quality assurance in colleges and universities: A new theoretical basis and test of management system

Abstract

Background

In response to the development of higher education quality management in the 1990s, institutions all over the world have set up equivalent internal teaching quality management systems based on local, regional, and global contexts. On the basis of the pertinent documents released by the Chinese Ministry of Education, this research also builds a teaching quality management system combined with neural networks.

Methods

GA-BP neural network and convolutional neural network are used to improve the evaluation method of teaching quality management and the monitoring method of classroom teaching quality.

Results

The improved results show that the accuracy of the upgraded GA-BP network for evaluating teaching quality is 89.62%, which is 11.45% higher than the baseline network. Convolutional neural network behavior recognition has high stability and repeatability, and the accuracy rate of classroom teaching quality supervision is 82.8% and the accuracy is 85.2%.

Discussion

GA-BP network optimizes the initial weight and structure of the neural network through genetic algorithm, and convolutional neural network provides a new technical path for the intelligent transformation of the internal quality management system of colleges and universities by monitoring the classroom behavior of teachers and students.