Academic Performance Prediction Using Machine Learning: A Comprehensive & Systematic Review
Praveena Chakrapani, D. Chitradevi
Abstract
Educational Institutions face numerous challenges today in providing quality and student-centric education to students. Despite the huge volume of data available with educational institutions, they lack a system that monitors and analyses students' performance in order to proactively take corrective actions that would channelize the efforts of the educational institution and the student, in the correct direction by adopting Student Intervention Strategies in a timely manner. With the advent of Educational Data Mining (EDM), there is a growing awareness amongst educational institutions to utilize Data Mining (DM) and Machine Learning (ML) techniques to analyze and predict the Academic Performance of students in a reliable, sophisticated and timely manner in order to ensure academic success for every student in its institution, irrespective of the student's academic caliber. The goal of this review paper is to present a comprehensive and systematic literature review of the numerous researches done in predicting students' performance through Machine Learning techniques and assess the quality of the accuracy of predictions in a clear and crisp manner. In this review, papers published during the period 2015 to 2022 through various leading publishers have been analyzed in-depth. This comprehensive and systematic analysis clearly reveals an increasing amount of research in the area of EDM for predicting Academic Performance of students, as well as, an increasing variety and combination of ML techniques and ensemble algorithms for accurate and timely Academic Performance Prediction (APP).