Building A Smart Academic Advising System Using Association Rule Mining

TL;DR

This paper presents a smart academic advising system that leverages association rule mining to recommend courses to students based on historical registration data, aiming to enhance advising efficiency and student performance.

Contribution

It introduces a novel application of association rule mining for personalized course advising in large universities, reducing advisor workload and improving student decision-making.

Findings

The system successfully generates relevant course association rules.

It improves the efficiency of academic advising processes.

Students receive tailored course recommendations based on historical data.

Abstract

In an academic environment, student advising is considered a paramount activity for both advisors and student to improve the academic performance of students. In universities of large numbers of students, advising is a time-consuming activity that may take a considerable effort of advisors and university administration in guiding students to complete their registration successfully and efficiently. Current systems are traditional and depend greatly on the effort of the advisor to find the best selection of courses to improve students performance. There is a need for a smart system that can advise a large number of students every semester. In this paper, we propose a smart system that uses association rule mining to help both students and advisors in selecting and prioritizing courses. The system helps students to improve their performance by suggesting courses that meet their current needs and at the same time improve their academic performance. The system uses association rule mining to find associations between courses that have been registered by students in many previous semesters. The system successfully generates a list of association rules that guide a particular student to select courses registered by similar students.