ABSTRACT
Implementing machine learning (ML) in the educational sector concerns extracting hidden information from the education dataset, which could prove useful in improving the education system by resolving Various factors which may result in the low grades of students, such as high level of alcohol and drug consumption, spending more time on extracurricular activities, lack of parents’ education, lack of proper motivation, and so on. In this paper, we used different ML approaches such as regression tree (RT), linear regression (LR), and random forest (RF) on an educational dataset of average grades obtained in Spanish and mathematics. In addition, we tried to analyze the primary factors that led to the high- and low-scoring students. For instance, our results demonstrate that although liquor utilization on weekdays and weekends is not the most grounded indicator of average student grades, variables like the ability to pursue higher education and the educational background of the mother have more impact on the grades.
