article · Measurement Interdisciplinary Research and Perspectives
This research evaluates predictive models for student performance using the Open University Learning Analytics Dataset. Several standard machine learning algorithms, including decision trees, random forests, and support vector machines, were tested alongside a deep neural network tuned with Bayesian Optimisation and HyperBand. To explain the model outputs, the system incorporates a hybrid feature scoring function and Shapley Additive Explanations. The tuned deep neural network achieved higher predictive accuracy than the baseline models, successfully classifying learners into distinction, pass, fail, and dropout outcomes. The explanatory framework identified student demographics, prior educational attainment, and engagement with online learning materials as key factors driving academic results.
Understanding why students struggle or drop out of higher education is essential for improving institutional support. By combining high predictive accuracy with interpretable explanations, this approach helps educators recognise which learner characteristics and study habits most strongly affect success, allowing institutions to target assistance more effectively.
The predictive framework could be integrated into educational technology platforms and university learning management systems to flag at-risk learners early. Prospective users include higher education administrators, course designers, and academic advisors. Based on the abstract, the research represents an applied and tested model evaluated on an open dataset, requiring further integration with live institutional data streams before full operational deployment.
AI-generated from the published abstract. Always read the original work before citing.
This study focuses on predicting student performance and identifying influential factors to optimize educational strategies, utilizing the Open University Learning Analytics Dataset (OULAD). After preprocessing the data, multiple machine learning models including KNN, Logistic Regression, Naive Bayes, SVM, Decision Trees, and Random Forests were evaluated. The core of our approach is an optimized Deep Neural Network (DNN) that employs Bayesian Optimization and HyperBand (BOHB) for hyperparameter tuning, significantly enhancing predictive accuracy. For robust feature analysis, a novel Hybrid Feature Scoring Function (HFSF) and an interpretable SHAP analysis were implemented on the optimized DNN to rank feature importance. The results demonstrate that the proposed BOHB-optimized DNN outperforms all baseline models in categorizing students into distinction, pass, fail, and dropout classes, while also identifying key influencing factors such as demographics, prior education, and online resource engagement. This work contributes an effective predictive framework for educational systems and highlights the potential of hybrid optimization techniques, with future research directed toward hybrid models and longitudinal data for improved accuracy and generalizability.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.1080/15366367.2026.2669152
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
