Student dropout remains a persistent challenge in higher education, particularly in developing countries where institutional support systems are often limited. This study develops a machine learning based predictive model for early identification of students at risk of dropping out within Nigerian universities. A quantitative research design was adopted using institutional data comprising academic performance, demographic characteristics, and behavioural indicators. The dataset was pre-processed through cleaning, encoding, and feature selection, and subsequently divided into training and testing subsets.Multiple classification algorithms including Logistic Regression, Decision Tree, Random Forest, Support Vector Machine, and K-Nearest Neighbors were implemented and evaluated using standard performance metrics such as accuracy, precision, recall, F1-score, and ROC-AUC. The results indicate that ensemble and kernel-based methods outperform traditional linear models, achieving accuracy levels exceeding 80% while significantly improving recall in identifying at-risk students. Key predictors of dropout include academic performance trends, attendance patterns, and student engagement indicators. The findings demonstrate the effectiveness of machine learning techniques in enabling early detection of student attrition risk. The study recommends the integration of interpretable predictive models into institutional information systems to support timely intervention strategies. Furthermore, it highlights the need for robust data governance frameworks to ensure ethical and sustainable deployment of predictive analytics in higher education.
Precious et al. (Fri,) studied this question.