Raqamli Transformatsiya va Sun’iy Intellekt
INTERPRETABLE MULTI-MODEL FUSION FOR ACADEMIC PERFORMANCE PREDICTION
1 “Tashkent Institute of Irrigation and Agricultural Mechanization Engineers” National Research University
Zenodo DOI: 10.5281/zenodo.23345413
Keywords
Abstract
Academic performance prediction is a core task in educational data mining and learning analytics, and its outcomes directly inform academic intervention decisions in higher education. A single model can hardly exploit multi-source, heterogeneous learning-behavior features; this study therefore proposes a multi-model fusion framework in which logistic regression, random forest, XGBoost, LightGBM and CatBoost serve as heterogeneous base learners, and decision-level fusion is realized through a stacking (logistic regression) meta-learner and an MLP (neural network) meta-learner. Stratified five-fold cross-validation on the public OULAD dataset (32,593 course-enrollment records, 46 features) shows that the best single model, CatBoost, attains an AUC of 0.9730, while both fusion strategies reach 0.9727, comparable to the best single model and achieving a better recall-precision balance. Ablation results show that assessment scores are the dominant information source, while virtual learning environment (VLE) behavioral features provide a stable incremental gain; SHAP analysis identifies the number of assessments attempted, mean score and VLE active days as the most important predictors. When strong single models are already near the performance ceiling, multi-model fusion thus adds value chiefly by reducing model-selection risk and by providing interpretable evidence for prediction.
References
- Li L. A Hybrid Ensemble Learning Framework for Educational Data Mining and Learning Analytics Research from an Interdisciplinary Perspective // Journal of Computing & Biomedical Informatics. 2026. Vol. 11. No. 2. DOI: 10.56979/1102/2026/1467.
- Zou W., Zhong W., Du J., Yuan L. Prediction of Student Academic Performance Utilizing a Multi-Model Fusion Approach in the Realm of Machine Learning // Applied Sciences. 2025. Vol. 15. No. 7. Pp. 3550. DOI: 10.3390/app15073550.
- Kelicha O., Blessing U. F., Anyanwu A., Nwagbara C. Ensemble Machine Learning for Student Performance Prediction in Higher Education: A Critical Review of Predictive Gains, Methodological Quality and Deployment Constraints // Asian Journal of Research in Computer Science. 2026. Vol. 19. DOI: 10.9734/ajrcos/2026/v19i9911.
- Okoye K., Nganji J. T., Escamilla J., Hosseini S. Machine Learning Model (RG-DMML) and Ensemble Algorithm for Prediction of Students' Retention and Graduation in Education // Computers and Education: Artificial Intelligence. 2024. Vol. 6. Pp. 100205. DOI: 10.1016/j.caeai.2024.100205.
- Wolpert D. H. Stacked generalization // Neural Networks. 1992. Vol. 5. No. 2. Pp. 241–259. DOI: 10.1016/S0893-6080(05)80023-1.
- Qiang M., Liu Z., Zhang R. Explainable AI in education: integrating educational domain knowledge into the deep learning model for improved student performance prediction // Scientific Reports. 2026. Vol. 16. No. 1. DOI: 10.1038/s41598-026-40538-y.
- Kalani N., Joshi H. An Improved Hybrid Stacking Ensemble Model for Predicting Student Academic Performance Using Machine Learning // International Journal for Research in Applied Science and Engineering Technology. 2026. Vol. 14. No. 3. Pp. 318–321. DOI: 10.22214/ijraset.2026.77793.
- Alhazmi E., Sheneamer A. Early predicting of students' performance in higher education // IEEE Access. 2023. Vol. 11. Pp. 27579–27589. DOI: 10.1109/ACCESS.2023.3250702.
- Omarbekova A., Nazyrova A., Bekmanova G., Lamasheva Z., Turebayeva R., Abzal Z. Comparative Evaluation of Machine Learning Models for Early Prediction of Student Academic Performance Using the Open University Learning Analytics Dataset // International Journal of Interactive Mobile Technologies (iJIM). 2026. Vol. 20. No. 17. DOI: 10.3991/ijim.v20i17.62574.
- Mahawar K., Rattan P. Empowering Education: Harnessing Ensemble Machine Learning Approach and ACO-DT Classifier for Early Student Academic Performance Prediction // Education and Information Technologies. 2025. Vol. 30. No. 4. Pp. 4639–4667. DOI: 10.1007/s10639-024-12976-6.
- Aldisa R. T., Rochim A. F., Triayudi A. Student Achievement Prediction Models: A PRISMA-Based Systematic Literature Review // Journal of Information Systems and Informatics. 2026. Vol. 8. No. 2. Pp. 2638–2663. DOI: 10.63158/journalisi.v8i2.152
- Lundberg S. M., Lee S.-I. A unified approach to interpreting model predictions // Advances in Neural Information Processing Systems 30. 2017. Pp. 4765–4774.
- Chen T., Guestrin C. XGBoost: a scalable tree boosting system // Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2016. Pp. 785–794. DOI: 10.1145/2939672.2939785.
- Abu Saa A., Al-Emran M., Shaalan K. Factors affecting students' performance in higher education: a systematic review of predictive data mining techniques // Technology, Knowledge and Learning. 2019. Vol. 24. Pp. 567–598. DOI: 10.1007/s10758-019-09408-7.
- Arévalo-Cordovilla F. E., Peña M. Evaluating ensemble models for fair and interpretable prediction in higher education using multimodal data // Scientific Reports. 2025. DOI: 10.1038/s41598-025-15388-9.
- Hermaliani E. H., Fanani A. Z., Santoso H. A., Affandy. An Explainable Stacking Ensemble Framework for Student Academic Performance Prediction in Learning Management Systems // Ingénierie des systèmes d information. 2026. Vol. 31. No. 4. Pp. 1251–1265. DOI:10.18280/isi.310420.
- Kuzilek J., Hlosta M., Zdrahal Z. Open University Learning Analytics dataset // Scientific Data. 2017. Vol. 4. Pp. 170171. DOI: 10.1038/sdata.2017.103.
- Breiman L. Random forests // Machine Learning. 2001. Vol. 45. No. 1. Pp. 5–32. DOI: 10.1023/A:1010933404324.
- Predicting student academic achievement using stacked ensemble learning with deep neural networks and fuzzy-based feature selection // Scientific Reports. 2025. DOI: 10.1038/s41598-025-20779-z.
- Lu Y., Rajendra A. S., Zhang J., Zhao T. Explainable Machine Learning for Student Performance Prediction // AI in Education. 2026. Vol. 2. No. 2. Pp. 17. DOI: 10.3390/aieduc2020017.