Application of Ensemble Classifier Techniques in Online Learning Performance Prediction

The COVID-19 pandemic has caused a global transition to online learning, presenting significant challenges in maintaining effective interaction and communication within educational environments. Traditional Educational Data Mining (EDM) methods often rely on single classifiers such as Decision Trees (DT), Naive Bayes (NB), and Support Vector Machines (SVM), which are widely used due to their simplicity and interpretability. However, these methods may lack the predictive accuracy and adaptability required in complex online learning contexts. To address this gap, this study explores the application of ensemble classifiers, including Bagging and Boosting techniques to enhance predictive performance. Results demonstrate that ensemble models, such as AdaBoost combined with SVM, achieve up to 82% accuracy, surpassing the performance reported in prior research using single classifiers. These findings suggest that ensemble classifiers offer a robust approach to personalized education in online learning, providing actionable insights to improve adaptive and flexible learning strategies. This research contributes to the growing body of knowledge on leveraging advanced EDM techniques to address the evolving needs of digital education.