This review explains how explainability transforms predictions into interventions that faculty and students can trust and focuses on higher-level metrics beyond simply describing accuracy, such as area under the curve (AUC), F1 score, fairness metrics, and long-term learning outcomes, which helps ensure that both the technical and educational effectiveness of the results are achieved.
Abstract
In higher education, the overlap between educational data mining (EDM) and explainable artificial intelligence (XAI) has increased in recent years, leading to a quest to balance performance, prediction, and explainability. At the institutional level, the accuracy of predictive models often exceeds 80%, but the practical use of predictions depends more on the interpretability and practicality of predictions than on statistical criteria. This review integrates two decades of research and explains how explainability transforms predictions into interventions that faculty and students can trust.
The benefit of this survey is to reduce the repetition of previous reviews. Therefore, this field is categorized in this research into six directions: performance prediction, risk detection in dropout, adaptive personalization and interaction analytics, multimodal natural language-based approaches, and obesity research with ethically based multicenter studies. The research brings together proven results, case studies, and conceptual contributions. The literature review focuses on less-researched techniques and methods, such as graph neural networks for relational inference, federated learning for privacy-preserving analysis, and large-scale text processing using generative language models. This research also aims to focus on higher-level metrics beyond simply describing accuracy, such as area under the curve (AUC), F1 score, fairness metrics, and long-term learning outcomes, which helps ensure that both the technical and educational effectiveness of the results are achieved. All of this suggests that interpretation and what it can do are not technicalities alone, but elements of the institutional trust between teachers in these institutions and their students. Case based evidence shows that local interpretable model-agnostic explanations LIME, Shapley additive explanations (SHAP), and other attention-based interpretations bridge between pedagogy and analytics rationalizing for the transparency of interventions.
The increasing availability of educational data has accelerated the application of machine learning (ML) to predict student outcomes and support data-driven educational decision-making. However, existing research varies considerably in algorithms, evaluation approaches, outcome variables, and geographical contexts. Thi...
Nomathemba Dolores Kutyauripo, Sibusisiwe Dube· International journal of res...· 0 citations
The research showed that using multiple data types significantly increases the predictive power of the model, and among all methods evaluated, multiple linear regression had the best balance between accuracy and computational requirement for predicting student performance.
Botan Onat, A. Bilge, A. Akın· International journal of 3d...· 0 citations
The Explainable Learning Analytics Framework (ELAF) as an integrated eight-layer conceptual architecture is suggested as an integrated eight-layer conceptual architecture to cover this persistent gap between the prediction accuracy of algorithms and the use of the algorithms in the educational domain.
Pratik Mishra, Nishesh Bishwas, Anjana Neupane· Apex Journal of Business and...· 0 citations
Student performance analysis is a prime component of educational data mining and
outcome-based education. Educational Data Mining and Machine Learning techniques have
received significant attention to predict student performance and support data-driven
decision-making in higher education. Previous work has applied arti...
R. Agnihotri, V. Sharma· ITEGAM- Journal of Engineeri...· 0 citations
As machine learning (ML) models are increasingly used to support decision making in higher education, there is a growing need to move beyond accurate prediction toward explanations that enable meaningful intervention. This paper examines counterfactual explanations (CFEs) as prescriptive complements to risk prediction...
D. H. Dinh, Yen Ngoc Nguyen, Ngọc Thị Hồng Trần· Advances in Online Education...· 0 citations
This narrative review examines how educational data mining (EDM) can inform instructional decisions about learning difficulties among novice programmers and focuses on target alignment, timing, and granularity.
Chui-Yang Meng· Applied and Computational En...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.