A carregar...
Publicação
Data-Driven Prediction of Student Success in Higher Education: A Predictive Modeling Study of Academic Outcomes via Binary Classification
| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 2.1 MB | Adobe PDF |
Orientador(es)
Resumo(s)
High dropout rates remain a critical challenge for higher education institutions, often resulting
in significant academic, social, and economic consequences. This study explores the use of
machine learning to develop predictive models capable of identifying students who are at risk
of academic failure early in their academic journey. Using a publicly available dataset from a
Portuguese higher education institution, the problem was reformulated as a binary
classification task, distinguishing between students who graduate and students who dropout.
After preprocessing and feature selection, six machine learning models were implemented
and evaluated: Logistic Regression, Decision Tree, Random Forest, Neural Network (with an
MLP architecture), Neural Network with Early Stopping, and XGBoost. The models were
assessed using accuracy, precision, recall, F1-score, ROC-AUC, and confusion matrix analysis.
Results showed that all models performed reasonably well, with the standard Neural Network
achieving the best overall balance between performance metrics and minimizing false
negatives, which were considered of particular importance given the imbalanced nature of
the dataset. The study highlights the potential of predictive analytics to enhance academic
support systems and offers insights into the importance of model selection criteria aligned
with institutional priorities and ethical considerations.
Descrição
Dissertation presented as the partial requirement for obtaining a Master's degree in Data Science and Advanced Analytics, specialization in Business Analytics
Palavras-chave
Student Success Prediction Machine Learning Educational Data Mining Early Warning Systems Higher Education Analytics Academic Dropout SDG 4 - Quality educa SDG 8 - Decent work and economic growth SDG 9 - Industry, innovation and infrastructure SDG 10 - Reduced inequalities