Research on Classroom Evaluation Algorithm Based on CNN Text Preprocessing

Abstract

Listen

Translate article

AbstractThe quality of classroom teaching is a key factor affecting the quality of higher education, and accurate classroom evaluation is an important means to improve the quality of classroom teaching. The data used for classroom evaluation generally include subjective evaluation unstructured text and objective structured data, but most classroom evaluations use objective structured data, and subjective evaluation of unstructured text has not been fully utilized. To solve this problem, a class evaluation back propagation neural network algorithm based on convolutional neural network text preprocessing is proposed. The algorithm uses CNN to convert the subjective data of the evaluation into structured data. Combined with the objective evaluation data and processed by the BP network, the emotional tendency of the overall evaluation quality of the classroom is obtained. Experiments show that the accuracy of sentiment classification with structured text data is 2.33% higher than that without adding text data. The addition of text data can more accurately evaluate the quality of classroom teaching.KeywordsConvolutional neural networkBackpropagation neural networkText preprocessingClassroom evaluation