Research on the Innovation of English Teaching Mode in Colleges and Universities by Integrating Deep Learning Algorithms

Abstract

In this paper, a new personalized recommendation system for English teaching resources is designed on the basis of deep learning, and a personalized English teaching model is innovatively proposed. The features of test information are continuously mined by convolutional neural network. Matrix decomposition technology and Bayesian criterion are adopted, so as to effectively characterize the association between users and English teaching resources, thus completing the recommendation of teaching resources. With the help of questionnaire data, the current status of the teaching practice of the Civics Smart Classroom in colleges and universities is understood. The test results show that the average test scores (P=0.001) and the degree of knowledge mastery (P=0.0045) of the experimental class and the control class are significantly different, indicating that the implementation of the personalized teaching mode based on the deep learning recommendation algorithm is effective. After the teaching practice, the students' independent learning, critical thinking, innovative thinking, problem solving, transfer thinking, effective communication, and cooperation ability are improved to different degrees.