Using Machine Learning Algorithm to Predict Student Pass Rates In Online Education

Abstract

In online education, the quality evaluation of education is crucial importance to schools and even to the entire range of educational institutions. There are many ways to evaluate online education. Taking the prediction of student pass rates as an example, many researchers have used machine Learning algorithms to predict student pass rates and find out important student features affecting learning. However, they did not establish feature model for online education that predicts the student pass rates and introduce deep neural network (DNN) algorithms - a new method in machine learning into online education. Therefore, this study first explores how to establish a feature model that predicts the student pass rates for online education, and then uses the grid search (GS) algorithm to optimize the decision tree algorithm (DT) and support vector machine (SVM) algorithm to improve the prediction accuracy. Finally, compared the improved algorithm with the DNN algorithm, we find a suitable algorithm for student pass rate prediction. The purpose of this study is to improve the quality of online teaching by predicting the student pass rates, increasing students' academic performance and strengthening online educational management.