Predicting Student Academic Performance using Machine Learning and Time Management Skill Data

Abstract

Prediction of student academic performance is an important aspect in the learning process. This study applies several machine learning models in predicting student academic performance using Time Management Skills data obtained from Time Structure Questionnaire (TSQ). Previously, some other data has been used as a feature in making predictions, but TSQ result had never been used before as a feature, even though it may shows the conditions of how students use their time in learning. Five different machine learning models were trained using TSQ data to predict student academic performance. In addition, student English performance is also predicted in the same way as a comparison As a result, the Linear Support Vector Machine model can predict student academic performance with 80% accuracy and English performance with 84% accuracy using TSQ data.