PREDICTING STUDENT GRADUATION CLASSIFICATIONS USING MACHINE LEARNING TECHNIQUES

Abstract

The ability to predict academic performance at the time of graduation is of profound importance to universities, especially in distinguishing factors affecting graduation ranking will contribute to improving the efficiency of student graduation ranking. This research uses many machine learning algorithms including K-nearest neighbor, Decision Tree, Random Forest, Logistic Regression, Support Vector Machine and Recurrent Neural Network to predict the graduation results of 1,817 full-time university students at Can Tho University of Technology including engineering and bachelor's programs from 2022 to 2024. The results showed that the Decision Tree model provided the most reliable predictions and fast training time. The factors affecting graduation classification included: GPA, age, major, gender, etc. Based on the experimental results, these factors were ranked to determine their impact on the graduation classification of students.