prediction of student performance using academic data and E-learning system usage features with ensemble learning approach

Abstract

This study aims to predict student academic performance by utilizing academic data and E-learning system usage features. In today’s digital era, data generated by E-learning systems holds significant potential for identifying factors that influence students' academic performance. Using an ensemble learning approach, this research combines multiple machine learning algorithms to enhance prediction accuracy. The data includes students' academic records, such as assignment and exam scores, along with interaction data within the E-learning platform. The results indicate that the ensemble learning approach achieves higher accuracy than single algorithms, making it a reliable predictive tool to support decision-making in educational environments.