Academic Advising for Online Higher Education Enrolment Based on Course Association Rules

Course enrolment is a critical aspect of a student's academic journey, and the ability to make informed decisions about which courses to take can greatly impact their educational success. A key factor that often goes unnoticed is the sequence in which courses are enrolled, which can significantly affect a student's ability to excel in their studies. This study delves into the relationship between course enrolment patterns and students’ performance. This paper presents an innovative approach that combines course enrolment association rule mining with the incorporation of grade score measures, aiming to provide a robust framework for informed academic decision support. Association rule mining is applied to identify patterns and relationships among the courses students have previously enrolled in. These patterns reveal the popularity of certain courses being chosen after the other. Interestingly enough, it has been observed that following a common enrolment pattern can cause a big drop or increase in different student groups. The extracted association rules not only assist with course recommendations but also enable academic advisors to caution students or prevent them from enrolling in courses where lower grades are predicted, particularly for students at risk of dropping out of studies. This paper presents a framework for improving the course selection process to advise students with different academic performances to enroll in a course or avoid it, benefiting both students and educational institutions. We used the dataset of Universitat Oberta de Catalunya on Informatics undergraduate studies for eight academic semesters to implement and evaluate the proposed method. The initial results show that integrating association rule mining with grade score measures in course enrolment decision-making offers a promising avenue for enhancing the quality of academic advising for course enrolment recommendations.