Personalized Course Recommendation System for Nigerian Secondary School Students Using Supervised Machine Learning Approach

Abstract

The high-rate diversity of courses offered in higher institutions has provided students with a broad spectrum of options and a desire for academic and career development. However, this abundance of choice has also introduced significant challenges in selecting courses that align with students' interests, skills, and long-term career goals. Traditional academic advisory systems which rely heavily on one-on one guidance from counselors or faculty, are constrained by the availability of advisors, the time required to provide tailored guidance, and the lack of data-driven insights into students' unique preferences and abilities. This paper presents a machine learning based personalized course recommendation system designed to assist students in selecting appropriate educational courses based on their Unified Tertiary Matriculation Examination (UTME) scores. Leveraging a comprehensive dataset of 1,000 students, the system employs advanced machine learning techniques, notably the XGBoost classifier, combined with Synthetic Minority Over-sampling Technique (SMOTE) to address class imbalance. Extensive feature engineering transforms raw examination scores and demographic variables into predictive features, enhancing model accuracy. The model was rigorously evaluated using stratified train-test splits and multiple performance metrics, achieving an overall accuracy exceeding 99%. Key insights include high predictive power of subject streams and individual subject scores in forecasting suitable courses for the students. Resulting recommendations provide actionable, interpretable guidance for students and counselors, facilitating informed decision-making and optimized academic pathways. This research demonstrates that machine learning models significantly enhance personalized learning experiences by effectively predicting suitable courses for students and also contributes a robust, datadriven methodology for educational planning support.