Unpacking the Black Box
A Hybrid XAI Framework for AutoGluon-Based Multiclass Student Outcome Prediction
DOI:
https://doi.org/10.7160/eriesj.2026.190103Keywords:
AutoGluon, Educational Data Mining, Explainable AI (XAI), Machine Learning, SDG 4, Student Dropout PredictionAbstract
High student dropout rates remain a significant impediment to achieving the United Nations SDG 4 (equitable education). While Artificial Intelligence (AI) offers robust early risk prediction, the intrinsic black-box nature of high-performing models constrains their transparency. This study designs and investigates a multi-layered Explainable AI (XAI)-based assessment framework to generate actionable insights for student retention. We utilized AutoGluon to construct high-performing multiclass classification models (Graduated, Dropout, or Enrolled) on a higher education dataset. To address the complexity of the AutoGluon-generated models, we employed a hybrid XAI framework that couples global interpretability via a decision tree surrogate model and local interpretability via LIME (Local Interpretable Model-agnostic Explanations). The analysis revealed that models from the Boosting family, particularly XGBoost with bagging level 2, achieved the highest predictive performance (exceeding 0.890 across all metrics). The global analysis demonstrated that academic factors were the primary drivers of prediction, but critical socio-economic factors, such as Tuition fees, also exerted significant influence. Local LIME analysis provided granular, case-specific insights, strongly linking dropout status to first-year academic challenges and to features such as age at enrollment. This integrated XAI approach transforms complex models into an interpretable system, supporting student retention and educational equity (SDG 4).
References
Alangari, N., El Bachir Menai, M., Mathkour, H. and Almosallam, I. (2023) ‘Exploring Evaluation Methods for Interpretable Machine Learning: A Survey’, Information, Vol. 14, No. 8, p. 469. https://doi.org/10.3390/info14080469
Apumayta, R. Q., Cayllahua, J. C., Pari, A. C., Choque, V. I., Valverde, J. C. C. and Ataypoma, D. H. (2024) ‘University dropout: A systematic review of the main determinant factors (2020–2024)’, F1000Research, Vol. 13, p. 253. https://doi.org/10.12688/f1000research.154263.2
Arjunan, G. (2021) ‘Implementing Explainable AI in Healthcare: Techniques for Interpretable Machine Learning Models in Clinical Decision-Making’, International Journal of Scientific Research and Management (IJSRM), Vol. 9, No. 05, pp. 597–603. https://doi.org/10.18535/ijsrm/v9i05.ec03
Arrieta, A.B., Díaz-Rodríguez, N., Del Ser, J., Bennetot, A., Tabik, S., Barbado, A., Garcia, S., Gil-Lopez, S., Molina, D., Benjamins, R., Chatila, R. and Herrera, F. (2020) ‘Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI’, Information Fusion, Vol. 58, No. 1, pp. 82–115. https://doi.org/10.1016/j.inffus.2019.12.012
Aulck, L., Velagapudi, N., Blumenstock, J. and West, J. (2016) ‘Predicting student dropout in higher education’, arXiv preprint, arXiv:1606.06364. https://doi.org/10.48550/arXiv.1606.06364
Awad, M. and Fraihat, S. (2023) ‘Recursive Feature Elimination with Cross-Validation with Decision Tree: Feature Selection Method for Machine Learning-Based Intrusion Detection Systems’, Journal of Sensor and Actuator Networks, Vol. 12, No. 5, p. 67. https://doi.org/10.3390/jsan12050067
Bandala, C. A. J. and Andrade, L. A. (2017) ‘Education, Poverty and the Trap of Poor Countries in the Face of Development’, Journal on Efficiency and Responsibility in Education and Science, Vol. 10, No. 4, pp. 101–108. https://doi.org/10.7160/eriesj.2017.100402
Berens, J., Schneider, K., Gortz, S., Oster, S. and Burghoff, J. (2019) ‘Early Detection of Students at Risk – Predicting Student Dropouts Using Administrative Student Data from German Universities and Machine Learning Methods’, Journal of Educational Data Mining, Vol. 11, No. 3, pp. 1–41. https://doi.org/10.5281/ZENODO.3594771
Birchard, K., Boccia, C., Lounder, H., Colston‐Nepali, L. and Friesen, V. (2025) ‘Popfinder: A Highly Effective Artificial Neural Network Package for Genetic Population Assignment’, Molecular Ecology Resources, Vol. 25, No. 1, p. e14096. https://doi.org/10.1111/1755-0998.14096
Blašková, V. and Staňková, M. (2023) ‘Graduate Employability as a Key to the Efficiency of Tertiary Education’, Journal on Efficiency and Responsibility in Education and Science, Vol. 16, No. 4, pp. 262–274. https://doi.org/10.7160/eriesj.2023.160401
Borisov, V., Leemann, T., Seßler, K., Haug, J., Pawelczyk, M. and Kasneci, G. (2024) ‘Deep Neural Networks and Tabular Data: A Survey’, IEEE Transactions on Neural Networks and Learning Systems, Vol. 35, No. 6, pp. 7499–7519. https://doi.org/10.1109/TNNLS.2022.3229161
ElShawi, R., Sherif, Y., Al‐Mallah, M. and Sakr, S. (2021) ‘Interpretability in healthcare: A comparative study of local machine learning interpretability techniques’, Computational Intelligence, Vol. 37, No. 4, pp. 1633–1650. https://doi.org/10.1111/coin.12410
Erickson, N., Mueller, J., Shirkov, A., Zhang, H., Larroy, P., Li, M. and Smola, A. (2020) ‘AutoGluon-Tabular: Robust and accurate AutoML for structured data’, arXiv preprint, arXiv:2003.06505. https://doi.org/10.48550/arXiv.2003.06505
Falvo, F. R. and Cannataro, M. (2024) ‘Explainability techniques for artificial intelligence models in medical diagnostic’, in: 2024 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), Lisbon: IEEE, pp. 6907–6913. https://doi.org/10.1109/BIBM62325.2024.10821826
Ferro, G. and D’Elia, V. (2020) ‘Higher Education Efficiency Frontier Analysis: A Review of Variables to Consider’, Journal on Efficiency and Responsibility in Education and Science, Vol. 13, No. 3, pp. 140–153. https://doi.org/10.7160/eriesj.2020.130304
Grinsztajn, L., Oyallon, E. and Varoquaux, G. (2022) ‘Why do tree-based models still outperform deep learning on typical tabular data?’, in: Advances in Neural Information Processing Systems (NeurIPS 2022), Vol. 35, pp. 507–520. https://doi.org/10.48550/arXiv.2207.08815
Guevara-Reyes, R., Ortiz-Garcés, I., Andrade, R., Cox-Riquetti, F. and Villegas-Ch, W. (2025) ‘Machine learning models for academic performance prediction: interpretability and application in educational decision-making’, Frontiers in Education, Vol. 10, p. 1632315. https://doi.org/10.3389/feduc.2025.1632315
Guidotti, R., Monreale, A., Ruggieri, S., Turini, F., Giannotti, F. and Pedreschi, D. (2019) ‘A Survey of Methods for Explaining Black Box Models’, ACM Computing Surveys, Vol. 51, No. 5, pp. 1–42. https://doi.org/10.1145/3236009
Ifenthaler, D. and Yau, J. Y-K. (2020) ‘Utilising learning analytics to support study success in higher education: a systematic review’, Educational Technology Research and Development, Vol. 68, No. 4, pp. 1961–1990. https://doi.org/10.1007/s11423-020-09788-z
Krüger, J. G. C., de Souza Britto Jr, A. and Barddal, J. P. (2023) ‘An explainable machine learning approach for student dropout prediction’, Expert Systems with Applications, Vol. 233, p. 120933. https://doi.org/10.1016/j.eswa.2023.120933
Kuhn, M. and Johnson, K. (2013) Applied Predictive Modeling, New York: Springer. https://doi.org/10.1007/978-1-4614-6849-3
Martins, M. V., Baptista, L., Machado, J. and Realinho, V. (2023) ‘Multi-Class Phased Prediction of Academic Performance and Dropout in Higher Education’, Applied Sciences, Vol. 13, No. 8, p. 4702. https://doi.org/10.3390/app13084702
Musaddiq, M. H., Sarfraz, M. S., Shafi, N., Maqsood, R., Azam, A. and Ahmad, M. (2022) ‘Predicting the Impact of Academic Key Factors and Spatial Behaviors on Students’ Performance’, Applied Sciences, Vol. 12, No. 19, p. 10112. https://doi.org/10.3390/app121910112
Nagy, M. and Molontay, R. (2024) ‘Interpretable Dropout Prediction: Towards XAI-Based Personalized Intervention’, International Journal of Artificial Intelligence in Education, vol. 34, pp. 274–300. https://doi.org/10.1007/s40593-023-00331-8
Nguyen, N. and Ngo, D. (2025) ‘Comparative analysis of boosting algorithms for predicting personal default’, Cogent Economics & Finance, Vol. 13, No. 1, p. 2465971. https://doi.org/10.1080/23322039.2025.2465971
OECD (2025) Education at a Glance 2025: OECD Indicators, Paris: OECD Publishing. https://doi.org/10.1787/1c0d9c79-en
Olive, U., Bosco, M. and Enan, N. (2025) ‘Predicting Student Dropout in Higher Education: An Ensemble Learning Approach with Feature Importance Analysis’, Journal of Information and Technology, Vol. 5, No. 4, pp. 31–40. https://doi.org/10.70619/vol5iss4pp31-40
Padmasiri, P. and Kasthuriarachchi, S. (2024) ‘Interpretable prediction of student dropout using explainable AI models’, in: 2024 International Research Conference on Smart Computing and Systems Engineering (SCSE), Colombo: IEEE, pp. 1–7. https://doi.org/10.1109/SCSE61872.2024.10550525
Panda, M. and Mahanta, S. R. (2023) ‘Explainable artificial intelligence for healthcare applications using random forest classifier with LIME and SHAP’, in: Balas, V. E., Kumar, R. and Srivastava, S. (eds.), Explainable, Interpretable, and Transparent AI Systems, Boca Raton: CRC Press, pp. 89–105. https://doi.org/10.1201/9781003442509-6
Realinho, V., Machado, J., Baptista, L. and Martins, M. V. (2022) ‘Predicting Student Dropout and Academic Success’, Data, Vol. 7, No. 11, p. 146. https://doi.org/10.3390/data7110146
Ribeiro, M. T., Singh, S. and Guestrin, C. (2016) ‘“Why should I trust you?”: Explaining the predictions of any classifier’, in: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD 2016), San Francisco: ACM, pp. 1135–1144. https://doi.org/10.1145/2939672.2939778
Sisman, S., Kara, A. and Aydinoglu, A. C. (2025) ‘Leveraging spatial data infrastructure for machine learning based building energy performance prediction’, PLOS One, Vol. 20, No. 1, p. e0335531. https://doi.org/10.1371/journal.pone.0335531
Villar, A. and de Andrade, C. R. V. (2024) ‘Supervised machine learning algorithms for predicting student dropout and academic success: a comparative study’, Discovery Artificial Intelligence, Vol. 4, No. 1, pp. 1–24. https://doi.org/10.1007/s44163-023-00079-z
Zanellati, A., Zingaro, S. P. and Gabbrielli, M. (2024) ‘Balancing performance and explainability in academic dropout prediction’, IEEE Transactions on Learning Technologies, Vol. 17, pp. 2086–2099. https://doi.org/10.1109/TLT.2024.3425959
Additional Files
Published
How to Cite
Issue
Section
Categories
License
Copyright (c) 2026 Marwan Nawae, Siripa Chankua, Massaya Longsaman
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors declare with this manuscript intended for publication to ERIES Journal that:
- all co-authors agree with the publication of the manuscript even after amendments arising from peer review;
- all co-authors agree with the posting of the full text of this work on the web page of ERIES Journal and to the inclusion of references in databases accessible on the internet;
- no results of other researchers were used in the submitted manuscript without their consent, proper citation, or acknowledgement of their cooperation or material provided;
- the results (or any part of them) used in the manuscript have not been sent for publication to any other journal nor have they already been published (or if so, that the relevant works are cited in this manuscript);
- submission of the manuscript for publication was completed in accordance with the publishing regulations pertaining to place of work;
- experiments performed comply with current laws and written consent of the Scientific Ethics Committee / National Animal Care Authority (as is mentioned in the manuscript submitted);
- grant holders confirm that they have been informed of the submitted manuscript and they agree to its publication.
Authors retain copyright and grant ERIES Journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the published work with an acknowledgement of its initial publication in ERIES Journal. Moreover, authors are able to post the published work in an institutional repository with an acknowledgement of its initial publication in ERIES Journal. In addition, authors are permitted and encouraged to post the published work online (e.g. institutional repositories or on their website) as it can lead to productive exchanges, as well as earlier and greater citation of published work.
