The Application of Three Machine Learning Algorithms in Student Performance Evaluation

Abstract

At present, the research of machine learning is a hot topic. In this paper, three machine learning algorithms, decision tree, support vector machine and random forest, are used to predict the students' achievement data sets. The results in the early stage of the data were analyzed to predict the average results in the later stage of the professional courses. The results show that the classification performance of the three classifier models is high, among which the random forest classifier is the best in the accuracy rate, precision rate, recall rate and F1 value. Moreover, the comprehensive forecast result and the course importance order can guide the student to carry on the pertinence remediation, and it's helpful for students to make specific explanations in class.