ENSEMBLE LEARNING IN EDUCATIONAL DATA ANALYSIS FOR IMPROVED PREDICTION OF STUDENT PERFORMANCE: A LITERATURE REVIEW

Abstract

The integration of advanced technology and digital platforms in modern education is essential for enhancing educational outcomes. Ensemble learning has emerged as a prominent approach in educational data analysis, demonstrating its effectiveness in improving student performance predictions. The study reviews the application of ensemble learning methods in educational data analysis to improve student performance prediction. The primary objective of this review is to highlight the effectiveness of ensemble approaches in achieving superior prediction accuracy compared to individual classifiers. Additionally, the review examines into the influence of feature selection techniques on optimizing ensemble models by identifying crucial attributes and mitigating the complexity of educational data. The findings show that ensemble learning offers a robust framework for tackling addressing  challenges in educational data mining, such as managing high-dimensional datasets and imbalanced classes. By incorporating feature selection methods, ensemble models become more efficient and scalable for various educational datasets. This review concludes that ensemble learning with integrated feature selection is a transformative tool for enhancing student performance prediction. Ensemble learning presents promising opportunities for driving innovation in educational data analysis and addressing the evolving challenges in education. The study offers valuable insights and benefits as a resource for educators, researchers and practitioners, shedding light on the transformative potential of ensemble learning in educational decision-making technology.