Student’s Academic Performance Improvement Using Adaptive Ensemble Learning Method

Abstract

Online learning platforms have transformed education by offering flexible, accessible, and interactive learning experiences. With advancements in technology and the increasing need for remote learning, these platforms empower students to study from anywhere at their own pace, offering various resources such as video lectures, assignments, quizzes, and discussion forums. These tools facilitate both self-paced learning and collaborative activities, allowing students to interact with peers, engage in group discussions, and work on joint projects. Big data analytics, in particular, plays a critical role in understanding student behaviour and cognitive processes, providing educators with valuable insights to personalize learning experiences more effectively. This study focuses on analysing student performance on online collaborative platforms through big data analytics, utilizing an ensemble model that integrates multiple Machine Learning (ML) algorithms to predict student outcomes more accurately. The proposed ensemble model achieved an accuracy of 98.87%, outperforming traditional classifiers in both accuracy and precision, particularly in identifying cognitive traits and predicting academic performance. These findings underscore the value of ensemble of ML in big data optimizing student engagement and success.