Home / Current Issue / Paper 1723819
Explainable AI-Driven Student Risk Prediction
Subject area: Science,Engineering and Technology · Area of research: Artificial Intelligence and Machine Learning
Abstract
Student attrition and academic failure remain persistent challenges in Indian higher education, where large cohorts make it difficult for advisors to identify struggling students early. Many machine-learning predictors operate as black boxes, which limits the trust, accountability and practical use that faculty and counselors can place in them. This paper proposes an explainable artificial intelligence (XAI)-driven framework for student risk prediction that combines ensemble classifiers (Random Forest and XGBoost) with SHAP and LIME explanations, a rule-based early warning layer and a human-in-the-loop intervention workflow. The framework brings academic, attendance and learning management system (LMS) data together in a modular, privacy-first architecture that includes role-based access control, tiered consent and periodic bias auditing. An evaluation protocol for the prediction pipeline is defined on the Open University Learning Analytics Dataset (OULAD), using only data available up to day 28 of a course. Every prediction is accompanied by a ranked list of contributing factors, so that advisors can respond to the reasons behind a risk estimate and not to the score alone. The framework offers a transparent, scalable and low-cost approach to the early identification and support of at-risk students.
Keywords
Algorithmic fairness, early warning system, explainable artificial intelligence, learning analytics, student risk prediction
References
[1] Romero C, Ventura S. Educational data mining and learning analytics: An updated survey. WIREs Data Min Knowl Discov. 2020;10(3):e1355. https://doi.org/10.1002/widm.1355
[2] Baker RSJd, Yacef K. The state of educational data mining in 2009: A review and future visions. J Educ Data Min. 2009;1(1):3–17.
[3] Slade S, Prinsloo P. Learning analytics: Ethical issues and dilemmas. Am Behav Sci. 2013;57(10):1510–1529. https://doi.org/10.1177/0002764213479366
[4] Lundberg SM, Lee SI. A unified approach to interpreting model predictions. In: Advances in Neural Information Processing Systems 30 (NeurIPS 2017); 2017. p. 4765–4774.
[5] Ribeiro MT, Singh S, Guestrin C. “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; 2016. p. 1135–1144. https://doi.org/10.1145/2939672.2939778
[6] Adadi A, Berrada M. Peeking inside the black-box: A survey on explainable artificial intelligence (XAI). IEEE Access. 2018;6:52138–52160. https://doi.org/10.1109/ACCESS.2018.2870052
[7] Barredo Arrieta A, Díaz-Rodríguez N, Del Ser J, Bennetot A, Tabik S, Barbado A, et al. Explainable artificial intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI. Inf Fusion. 2020;58:82–115. https://doi.org/10.1016/j.inffus.2019.12.012
[8] Hellas A, Ihantola P, Petersen A, Ajanovski VV, Gutica M, Hynninen T, et al. Predicting academic performance: A systematic literature review. In: Proceedings Companion of the 23rd Annual ACM Conference on Innovation and Technology in Computer Science Education (ITiCSE 2018 Companion); 2018. p. 175–199. https://doi.org/10.1145/3293881.3295783
[9] Arnold KE, Pistilli MD. Course Signals at Purdue: Using learning analytics to increase student success. In: Proceedings of the 2nd International Conference on Learning Analytics and Knowledge (LAK ’12); 2012. p. 267–270. https://doi.org/10.1145/2330601.2330666
[10] Viberg O, Hatakka M, Bälter O, Mavroudi A. The current landscape of learning analytics in higher education. Comput Human Behav. 2018;89:98–110. https://doi.org/10.1016/j.chb.2018.07.027
[11] Mehrabi N, Morstatter F, Saxena N, Lerman K, Galstyan A. A survey on bias and fairness in machine learning. ACM Comput Surv. 2021;54(6):115. https://doi.org/10.1145/3457607
[12] Hardt M, Price E, Srebro N. Equality of opportunity in supervised learning. In: Advances in Neural Information Processing Systems 29 (NeurIPS 2016); 2016. p. 3315–3323.
[13] Government of India. The Digital Personal Data Protection Act, 2023 (Act No. 22 of 2023). New Delhi: Ministry of Electronics and Information Technology; 2023.
[14] Breiman L. Random forests. Mach Learn. 2001;45(1):5–32. https://doi.org/10.1023/A:1010933404324
[15] Chen T, Guestrin C. XGBoost: A scalable tree boosting system. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining; 2016. p. 785–794. https://doi.org/10.1145/2939672.2939785
[16] Kuzilek J, Hlosta M, Zdrahal Z. Open University Learning Analytics dataset. Sci Data. 2017;4:170171. https://doi.org/10.1038/sdata.2017.171
[17] Cortez P, Silva A. Using data mining to predict secondary school student performance. In: Proceedings of the 5th Future Business Technology Conference (FUBUTEC 2008); 2008. p. 5–12.
How to cite this paper
@article{1723819,
author = {Priyanga M, Ranjith V, Pranesh B, Sankar A, Karthikeyan R},
title = {Explainable AI-Driven Student Risk Prediction},
journal = {Iconic Research And Engineering Journals},
year = {2026},
volume = {10},
number = {4},
pages = {5306-5318},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1723819.pdf},
abstract = {Student attrition and academic failure remain persistent challenges in Indian higher education, where large cohorts make it difficult for advisors to identify struggling students early. Many machine-learning predictors operate as black boxes, which limits the trust, accountability and practical use that faculty and counselors can place in them. This paper proposes an explainable artificial intelligence (XAI)-driven framework for student risk prediction that combines ensemble classifiers (Random Forest and XGBoost) with SHAP and LIME explanations, a rule-based early warning layer and a human-in-the-loop intervention workflow. The framework brings academic, attendance and learning management system (LMS) data together in a modular, privacy-first architecture that includes role-based access control, tiered consent and periodic bias auditing. An evaluation protocol for the prediction pipeline is defined on the Open University Learning Analytics Dataset (OULAD), using only data available up to day 28 of a course. Every prediction is accompanied by a ranked list of contributing factors, so that advisors can respond to the reasons behind a risk estimate and not to the score alone. The framework offers a transparent, scalable and low-cost approach to the early identification and support of at-risk students.},
keywords = {Algorithmic fairness, early warning system, explainable artificial intelligence, learning analytics, student risk prediction},
month = {October},
}