Analysis and Prediction of CET4 Score Based on RNN

Educational data mining technology is an application technology to improve and assist teaching practice and educational research and it is gradually concerned by researchers at home and abroad. However, the research of educational data mining in China started late, especially the research based on recurrent neural network and other deep learning algorithms. In recent years, recurrent neural network introduces a feedback mechanism in the hidden layer to realize the effective information analysis and data processing. Recurrent neural network has become a research hotspot in the fields of classification and prediction, natural language processing, computer vision and so on. Firstly, this paper makes a correlation analysis of College English Test Band 4 (CET4) scores and students' basic information for 7367 undergraduates in a Shanghai university. And then this paper realizes the prediction of CET4 scores based on students' basic information by recurrent neural network. It is an effective of recurrent neural network in educational data mining. The prediction results will provide guidance for graded teaching of college English.