Key points are not available for this paper at this time.
The exploration of machine learning applications for predicting students' academic performance has become a pivotal research domain. This systematic literature review screened 83 indexed research articles between 2020-2023 that examined machine learning applications in predicting academic achievement. The findings highlighted that ensemble learning outperformed other methods in predicting academic performance, achieving an average accuracy rate of 87.67%. This was closely followed by the support vector machine (SVM) approach, which attained an average accuracy of 84.30%. Demographic, academic, and behavioral factors were found to be significant predictors of academic achievement. This research emphasized the significance of early identification of students with issues and the necessity for timely interventions to enhance their educational outcomes. Predictive factors are essential as they can assist educators and policymakers in figuring out students who may be at risk of underperformance and in providing targeted interventions to enhance the quality of educational outcomes. This is in line with SDG 4, which focuses on quality education.
Wu et al. (Wed,) studied this question.