Abstract

Predicting student dropout in online learning platforms such as MOOCs and institutional LMS platforms, is a critical challenge in educational data mining. Although numerous machine learning models have been proposed to predict dropout likelihood, the lack of model interpretability has limited their practical deployment in educational settings. This paper proposes a stacked ensemble machine learning model combining Logistic Regression, Random Forest, and XGBoost, with explainable AI techniques to identify at-risk learners using behavioral and demographic features. The dataset, obtained from Kaggle’s MOOC Dropout Prediction challenge, was cleaned, balanced, and subjected to feature selection to prevent information leakage. With SHAP interpretability, the model achieves an accuracy of 65%, ROC AUC of 0.71, and PR AUC of 0.73. Our results show that dropout prediction is feasible using early behavioral data, and stacked models offer a promising balance of performance and transparency. This work contributes a replicable, explainable architecture suitable for real-time educational intervention systems.

References

• M. Jeon, S. Kim, and J. Kim, “Dropout prediction over weeks in MOOCs via interpretable multi-layer representation learning,” arXiv preprint arXiv: 2002.01598, 2020.
• R. Swamy, A. Sinha, and K. J. Shipp, “Evaluating the explainers: Black-box explainable machine learning for student success prediction in MOOCs,” arXiv preprint arXiv: 2207.00551, 2022.
• S. Krüger, J. E. del Río, and A. Ortega, “An explainable machine learning approach for student dropout prediction,” *Expert Systems with Applications*, vol. 228, 2023, Art. no. 120338.
• G. Dekker, M. Pechenizkiy, and J. Vleeshouwers, “Predicting students drop out: A case study,” in *Proc. International Conference on Educational Data Mining (EDM)*, 2009, pp. 41–50.
• F. Marcolino, L. R. de Lima, and A. dos Santos, “Student dropout prediction through machine learning optimization: insights from Moodle log data,” *Scientific Reports*, vol. 15, no. 1, 2025, Art. no. 1.
• L. Lamsiyah, M. Lakhouaja, M. Quafafou, and S. El Ghazi, “Privacy-preserving federated learning for student dropout prediction: Enhancing model transparency with explainable AI,” in *Advances in Knowledge Discovery and Data Mining*, Springer, 2025, pp. 435–446.
• I. Elbouknify et al., “AI-based identification and support of at-risk students: A case study of the Moroccan education system,” arXiv preprint arXiv: 2504.07160, 2025.
• A. L. Jimenez Martinez, K. Sood, and R. Mahto, “Early detection of at-risk students using machine learning,” arXiv preprint arXiv: 2412.09483, 2024.
• S. Albugami, H. Almaghrabi, and A. Wali, “From data to decision: Machine learning and explainable AI in student dropout prediction,” *Journal of e-Learning and Higher Education*, vol. 2024, Article ID 246301.
• H. Cigdem and O. Yildirim, “Effects of students’ characteristics on online learning readiness: A vocational college example,” *Turkish Online Journal of Distance Education*, vol. 15, no. 3, pp. 80–93, 2014.
• “Predicting student outcomes in online courses using machine learning techniques: A review,” *Sustainability*, vol. 14, no. 10, 2022.
• S. Ardchir et al., “Improving prediction of MOOCs student dropout using a feature engineering approach,” in *AI2SD*, Springer, 2020, pp. 150–161.
• P. Ghamisi and J. A. Benediktsson, “Dropout prediction model in MOOC based on clickstream data and student sample weight,” *Soft Computing*, 2021.
• J. Gardner and C. Brooks, “Student success prediction in MOOCs,” *User Modeling and User-Adapted Interaction*, vol. 28, no. 2, pp. 127–203, 2018.
• “A survey of machine learning approaches and techniques for student dropout prediction,” *Data Science Journal*, 2019.