Using Learning Analytics to Improve Students' Enrollments in Higher Education

Abstract

In the last years there has been a growing interest in adopting learning analytics (LA) in higher and further education systems. LA assists the institutional stakeholders to enhance the learning process, ameliorate the teaching activities, make adequate decisions and take appropriate actions based on the collection, analysis, and reporting of data generated from individual learners. The learning analytics approach aims to achieve many objectives, one of them is prediction which is the center of this research. In this paper, we conduct a comparative study between three machine learning algorithms; Decision Tree (DT), Random Forest (RF), and Support Vector Machine (SVM); to predict the stream of new enrollments in the first year of higher education. As a case study, the predictive model is applied to new enrollments in the first year of the University Diploma of Technology (DUT) at the Higher School of Technology in Meknes, Morocco (ESTM). The performance of the classifiers is tested with and without the use of SMOTE data balancing on a dataset of 53554 students collected between 2016 and 2019. The obtained results show the best algorithm to predict the most accurate enrollments of students.