Explainable machine learning with class balancing for predicting student academic performance

(1) * Zamtinah Zamtinah Mail (Universitas Negeri Yogyakarta, Indonesia)
(2) Rianto Rianto Mail (University of Technology Yogyakarta, Indonesia)
(3) Paulus Insap Santosa Mail (Universitas Gadjah Mada, Indonesia)
*corresponding author

Abstract


Accurate prediction of student academic performance is crucial for implementing effective early intervention strategies in higher education. However, existing models often struggle with class imbalance and lack interpretability, limiting their practical adoption. This methodological proof-of-concept develops a robust predictive framework by benchmarking multiple machine learning algorithms, specifically Logistic Regression, Decision Tree, and the standard Random Forest, against a proposed Random Forest classifier with class balancing. Utilizing a public simulation dataset of 2,392 student records, the proposed model incorporates algorithmic penalties for misclassifications of the minority class to address data skewness. Experimental results demonstrate that this balanced approach significantly outperforms the linear and standard tree-based baselines, achieving an Accuracy of 92.3% and an F1 Macro score of 0.872. The results confirm the model's superior capability in identifying at-risk students without compromising overall precision. Furthermore, explainability analysis using SHAP (Shapley Additive exPlanations) identifies Grade Point Average (GPA) as the dominant predictor, while highlighting Absences and Weekly Study Time as critical behavioral leading indicators. These findings demonstrate that integrating predictive accuracy with explainable AI supports proactive and evidence-based academic governance. Future research is recommended to validate this framework using longitudinal data streams within Learning Management Systems, enabling real-time monitoring.

Keywords


Student Academic Performance; Explainable Machine Learning; Class Imbalance; SHAP; Early Intervention

   

DOI

https://doi.org/10.26555/ijain.v12i3.2379
      

Article metrics

Abstract views : 208 | PDF views : 3

   

Cite

   

Full Text

Download

References


[1] S. Ghimire et al., “Explainable artificial intelligence-machine learning models to estimate overall scores in tertiary preparatory general science course,” Comput. Educ. Artif. Intell., vol. 7, p. 100331, 2024, doi: 10.1016/j.caeai.2024.100331.

[2] F. da Conceição Silva, A. M. Santana, and R. M. Feitosa, “An investigation into dropout indicators in secondary technical education using explainable artificial intelligence,” IEEE Rev. Iberoam. Tecnol. del Aprendiz., vol. 20, pp. 105–114, 2025, doi: 10.1109/RITA.2025.3566095.

[3] W.-C. Choi, C.-T. Lam, and A. J. Mendes, “Enhance learning performance predictions with explainable machine learning,” in 2024 IEEE Frontiers in Education Conference (FIE), 2024, pp. 1–9, doi: 10.1109/FIE61694.2024.10893036.

[4] G. A. Korsah, J. K. Adjei, J. A. Mensah, L. Arthur, A. S. Ariff, and R. Awuah, “The impact of artificial intelligence on higher education: Perspectives and insights from computer science faculty and students in Ghana,” in Proceedings of the ACM Global Computing Education Conference 2025-Volume 1, 2025, pp. 85–91, doi: 10.1145/3736181.3747154.

[5] M. Hoq, P. Brusilovsky, and B. Akram, “Explaining explainability: Early performance prediction with student programming pattern profiling,” J. Educ. Data Min., vol. 16, no. 2, pp. 115–148, 2024.

[6] S. Aswal, N. J. Ahuja, and R. Mehra, “Handling imbalance and limited data in thyroid ultrasound and diabetic retinopathy datasets using discrete levy flights grey wolf optimizer based random forest for robust medical data classification,” ACM Trans. Asian Low-Resource Lang. Inf. Process., 2024, doi: 10.1145/3648363.

[7] I. Arshad et al., “Performance of classification algorithms under class imbalance: Simulation and real-world evidence,” IEEE Access, 2025, doi: 10.1109/ACCESS.2025.3620264.

[8] M. M. Islam, F. H. Sojib, M. F. H. Mihad, M. Hasan, and M. Rahman, “The integration of explainable AI in Educational Data Mining for student academic performance prediction and support system,” Telemat. Informatics Reports, vol. 18, p. 100203, 2025, doi: 10.1016/j.teler.2025.100203.

[9] M. Altalhan, A. Algarni, and M. T.-H. Alouane, “Imbalanced data problem in machine learning: A review,” IEEE access, vol. 13, pp. 13686–13699, 2025, doi: 10.1109/ACCESS.2025.3531662.

[10] A. Ekinci and S. Sen, “Forecasting bank failure in the US: A cost-sensitive approach,” Comput. Econ., vol. 64, no. 6, pp. 3161–3179, 2024, doi: 10.1007/s10614-023-10537-6.

[11] J. A. Ali, A. H. Muse, M. K. Abdi, T. A. Ali, Y. H. Muse, and M. A. Cumar, “Machine learning-driven analysis of academic performance determinants: Geographic, socio-demographic, and subject-specific influences in Somaliland’s 2022–2023 national primary examinations,” Int. J. Educ. Res. Open, vol. 8, p. 100426, 2025, doi: 10.1016/j.ijedro.2024.100426.

[12] H. Mastour, T. Dehghani, E. Moradi, and S. Eslami, “Explainable artificial intelligence for predicting medical students’ performance in comprehensive assessments,” Sci. Rep., vol. 15, no. 1, p. 23752, 2025, doi: 10.1038/s41598-025-07460-1.

[13] R. Goran et al., “Identifying and understanding student dropouts using metaheuristic optimized classifiers and explainable artificial intelligence techniques,” IEEE Access, vol. 12, pp. 122377–122400, 2024, doi: 10.1109/ACCESS.2024.3446653.

[14] A. Zanellati, S. P. Zingaro, and M. Gabbrielli, “Balancing performance and explainability in academic dropout prediction,” IEEE Trans. Learn. Technol., vol. 17, pp. 2086–2099, 2024, doi: 10.1109/TLT.2024.3425959.

[15] N. R. Raji, R. M. S. Kumar, and C. L. Biji, “Explainable machine learning prediction for the academic performance of deaf scholars,” Ieee Access, vol. 12, pp. 23595–23612, 2024, doi: 10.1109/ACCESS.2024.3363634.

[16] A. Kala, O. Torkul, T. T. Yildiz, and I. H. Selvi, “Early prediction of student performance in face-to-face education environments: A hybrid deep learning approach with XAI techniques,” IEEE Access, vol. 12, pp. 191635–191649, 2024.

[17] M. Abdalkareem and N. Min-Allah, “Explainable models for predicting academic pathways for high school students in Saudi Arabia,” Ieee Access, vol. 12, pp. 30604–30626, 2024, doi: 10.1109/ACCESS.2024.3369586.

[18] S. Mustofa, Y. R. Emon, S. Bin Mamun, S. A. Akhy, and M. T. Ahad, “A novel AI-driven model for student dropout risk analysis with explainable AI insights,” Comput. Educ. Artif. Intell., vol. 8, p. 100352, 2025, doi: 10.1016/j.caeai.2024.100352.

[19] F. T. Johora, M. N. Hasan, A. Rajbongshi, M. Ashrafuzzaman, and F. Akter, “An explainable AI-based approach for predicting undergraduate students academic performance,” Array, vol. 26, p. 100384, 2025, doi: 10.1016/j.array.2025.100384.

[20] N. Farhah, M. Adnan, A. A. Alqarni, M. I. Uddin, and T. H. H. Aldhyani, “AI-driven innovation using multimodal and personalized adaptive education for students with special needs,” IEEE Access, vol. 13, pp. 69790–69820, 2025, doi: 10.1109/ACCESS.2025.3560863.

[21] O. Embarak, “A behaviour-driven framework for smart education: Leveraging explainable AI and IoB in personalized learning systems,” Procedia Comput. Sci., vol. 265, pp. 457–466, 2025, doi: 10.1016/j.procs.2025.07.205.

[22] P. Buñay-Guisñan, J. A. Lara, A. Cano, R. Cerezo, and C. Romero, “Easing the Prediction of Student Dropout for everyone integration AutoML and Explainable Artificial Intelligence,” in Proceedings of the 17th International Conference on Educational Data Mining, 2024, pp. 857–861.

[23] M. Saqr, A. Tlili, and S. López-Pernas, “Automating individualized machine learning and AI prediction using AutoML: the case of idiographic predictions,” in Advanced Learning Analytics Methods: AI, Precision and Complexity, Springer, 2025, pp. 565–593, doi: 10.1007/978-3-031-95365-1_22.

[24] M. Lünich and B. Keller, “Explainable artificial intelligence for academic performance prediction. an experimental study on the impact of accuracy and simplicity of decision trees on causability and fairness perceptions,” in Proceedings of the 2024 ACM Conference on Fairness, Accountability, and Transparency, 2024, pp. 1031–1042, doi: 10.1145/3630106.3658953.

[25] F. Chak-Him and N. G. Siu-Ping, “AI-driven simulation-based research: A new frontier in research innovation,” Comput. Educ. Artif. Intell., p. 100453, 2025, doi: 10.1016/j.caeai.2025.100453.

[26] S. Golazad, A. Mohammadi, A. Rashidi, and M. Ilbeigi, “From raw to refined: Data preprocessing for construction machine learning (ML), deep learning (DL), and reinforcement learning (RL) models,” Autom. Constr., vol. 168, p. 105844, 2024, doi: 10.1016/j.autcon.2024.105844.

[27] B. Boukebous, J.-E. Galimard, and D. Biau, “Improving Robustness and Interpretability of Models Including Machine Learning in Orthopedics through Data Preprocessing,” Orthopaedics & Traumatology: Surgery & Research. Elsevier, p. 104390, 2025, doi: 10.1016/j.otsr.2025.104390.

[28] M. Razavi et al., “Machine learning, deep learning, and data preprocessing techniques for detecting, predicting, and monitoring stress and stress-related mental disorders: scoping review,” JMIR Ment. Heal., vol. 11, no. 1, p. e53714, 2024, doi: 10.2196/53714.

[29] K. N. Savitha, L. Pinto, S. Gopalan, and P. Balasubramaniam, “Equivalence class and modified Gaussian methods for normalization of time series data on AI models,” Expert Syst. Appl., vol. 277, p. 127166, 2025, doi: 10.1016/j.eswa.2025.127166.

[30] M. Chen and Z. Liu, “Predicting performance of students by optimizing tree components of random forest using genetic algorithm,” Heliyon, vol. 10, no. 12, 2024, doi: 10.1016/j.heliyon.2024.e32570.

[31] M. Nachouki and M. Abou Naaj, “Predicting student performance to improve academic advising using the random forest algorithm,” Int. J. Distance Educ. Technol., vol. 20, no. 1, pp. 1–17, 2022, doi: 10.4018/IJDET.296702.

[32] T. Wahyuningsih, D. Manongga, I. Sembiring, and S. Wijono, “Comparison of effectiveness of logistic regression, naive bayes, and random forest algorithms in predicting student arguments,” Procedia Comput. Sci., vol. 234, pp. 349–356, 2024, doi: 10.1016/j.procs.2024.03.014.

[33] H. K. Chinmayi, K. C. Flynn, G. S. Baath, P. Gowda, B. Northup, and A. Ashworth, “Monitoring legume nutrition with machine learning: The impact of splits in training and testing data,” Appl. Soft Comput., vol. 176, p. 113186, 2025, doi: 10.1016/j.asoc.2025.113186.

[34] L. Geißel and P. Wiederkehr, “A generalized framework for predicting process robustness in the context of machining aero engine components,” Procedia CIRP, vol. 133, pp. 466–470, 2025, doi: 10.1016/j.procir.2025.02.080.

[35] I. Z. Sadiq, F. S. Abubakar, M. A. Saliu, B. S. Katsayal, A. Salihu, and A. Muhammad, “Machine learning algorithms for predictive modeling of dyslipidemia-associated cardiovascular disease risk in pregnancy: a comparison of boosting, random forest, and decision tree regression,” Bull. Natl. Res. Cent., vol. 49, no. 1, p. 6, 2025, doi: 10.1186/s42269-024-01295-y.

[36] R. Iranzad and X. Liu, “A review of random forest-based feature selection methods for data science education and applications,” Int. J. Data Sci. Anal., vol. 20, no. 2, pp. 197–211, 2025, doi: 10.1007/s41060-024-00509-w.

[37] S. Álvarez-Diez, J. S. Baixauli-Soler, G. Lozano-Reina, and D. R.-L. Rey, “Subsidies for investing in energy efficiency measures: Applying a random forest model for unbalanced samples,” Appl. Energy, vol. 359, p. 122725, 2024, doi: 10.1016/j.apenergy.2024.122725.

[38] B. Elkari, Y. Chaibi, and T. Kousksou, “Random forest with feature selection and K-fold cross validation for predicting the electrical and thermal efficiencies of air based photovoltaic-thermal systems,” Energy Reports, vol. 12, pp. 988–999, 2024, doi: 10.1016/j.egyr.2024.07.002.

[39] L. Zhang, Z. Luo, B. Su, Z. Song, J. Deng, and X. Ji, “Prediction of explosion hazard of aluminum powder two-phase mixed system using random forest based on K-fold cross-validation,” J. Loss Prev. Process Ind., vol. 94, p. 105574, 2025, doi: 10.1016/j.jlp.2025.105574.

[40] A. Shebl, D. Abriha, M. Dawoud, M. A. H. Ali, and Á. Csámer, “PRISMA vs. Landsat 9 in lithological mapping− a K-fold Cross-Validation implementation with Random Forest,” Egypt. J. Remote Sens. Sp. Sci., vol. 27, no. 3, pp. 577–596, 2024, doi: 10.1016/j.ejrs.2024.07.003.

[41] D. Pawade, D. Bakhai, T. Admane, R. Arya, Y. Salunke, and Y. Pawade, “Evaluating the Performance of Different Machine Learning Models for Metabolic Syndrome Prediction,” Procedia Comput. Sci., vol. 235, pp. 2932–2941, 2024, doi: 10.1016/j.procs.2024.04.277.

[42] S. Wali, Y. A. Farrukh, and I. Khan, “Explainable AI and random forest based reliable intrusion detection system,” Comput. Secur., vol. 157, p. 104542, 2025, doi: 10.1016/j.cose.2025.104542.

[43] W. C. Choi, C.-T. Lam, and A. J. Mendes, “Analyzing the interpretability of machine learning prediction on student performance using shapley additive explanations,” in 2024 IEEE International Conference on Teaching, Assessment and Learning for Engineering (TALE), 2024, pp. 1–8, doi: 10.1109/TALE62452.2024.10834292.

[44] D. Hooshyar and Y. Yang, “Problems with SHAP and LIME in interpretable AI for education: A comparative study of post-hoc explanations and neural-symbolic rule extraction,” IEEE Access, vol. 12, pp. 137472–137490, 2024, doi: 10.1109/ACCESS.2024.3463948.

[45] R. Guevara-Reyes, I. Ortiz-Garcés, R. Andrade, F. Cox-Riquetti, and W. Villegas-Ch, “Machine learning models for academic performance prediction: Interpretability and application in educational decision-making,” in Frontiers in Education, 2025, vol. 10, p. 1632315, doi: 10.3389/feduc.2025.1632315.

[46] A. Bansal, R. Sharma, and M. Kathuria, “Diversity-regularized multi-objective optimized Random Forest for fall detection using cost-sensitive and key point feature-based learning,” Comput. Electr. Eng., vol. 128, p. 110684, 2025, doi: 10.1016/j.compeleceng.2025.110684.

[47] R. M. Santos and R. Henriques, “Accurate, timely, and portable: Course-agnostic early prediction of student performance from LMS logs,” Comput. Educ. Artif. Intell., vol. 5, p. 100175, 2023, doi: 10.1016/j.caeai.2023.100175.




Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

___________________________________________________________
International Journal of Advances in Intelligent Informatics
ISSN 2442-6571  (print) | 2548-3161 (online)
Organized by UAD and ASCEE Computer Society
Published by Universitas Ahmad Dahlan
W: http://ijain.org
E: info@ijain.org (paper handling issues)
 andri.pranolo.id@ieee.org (publication issues)

View IJAIN Stats

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0