Developing a predictive model for student academic performance using machine learning techniques

View Full Paper

Ask AI

Bookmark

Authors

Discussion

Analysis reveals machine learning models improve forecast accuracy in student academic performance, highlighting equity concerns.

XGBoost outperformed traditional models, achieving R² of .70, demonstrating superior prediction in diverse student populations.
Feature importance analysis indicated demographic factors like Students with Disabilities significantly impacted predictions.
Ensemble methods like Random Forest and XGBoost effectively captured complex interactions within educational data structures.
Recommendations emphasize fairness in model deployment and the necessity for equitable educational interventions.

Bello et al. (2025) studied this question.

View Full Paper

Ask AI

Bookmark