← Back⌂ All journals
Review of Artificial Intelligence in Educatione-ISSN 2965-4688v. 7 (2026)Open Access1 min read

Frugal Artificial Intelligence for Educational Early-Warning Systems: A Constrained Feature-Set Modeling Framework for Resource-Constrained Environments

  1. MZInstituto Superior de Transportes e Comunicações (ISUTC) Maputo, Mozambique
Read as flipbookAll articles in this issueCite this article

Abstract

A inteligência artificial (IA) é cada vez mais aplicada à previsão educacional, à analítica de aprendizagem e à tomada de decisões institucionais. No entanto, muitos sistemas baseados em IA pressupõem conectividade contínua e infraestrutura em nuvem, limitando a sua implementação em ambientes com restrições de recursos. Este estudo propõe e avalia o framework Constrained Feature-Set Modeling (CFSM), uma arquitetura de IA frugal concebida para sistemas de alerta precoce educacionais sob restrições computacionais e de conectividade. Adotámos uma abordagem de Design Science Research e utilizámos um benchmark sintético contendo 12.450 perfis de estudantes. O CFSM utiliza três indicadores comportamentais — Tendência de Assiduidade, Frequência de Entrega de Tarefas e Desempenho Académico Ponderado — com um classificador Random Forest restrito composto por 50 árvores e uma profundidade máxima de 10. O framework alcançou 90,0% de exatidão (accuracy), 69,2% de precisão (precision), 92,1% de sensibilidade (recall), 79,0% de F1-score e 0,941 de AUC-ROC. O modelo requereu aproximadamente 1,8 MB de armazenamento e 42 MB de pico de memória RAM em hardware legado. O CFSM suportou com sucesso a inferência local sem necessidade de conectividade de rede contínua. Os resultados sugerem que a IA frugal oferece uma alternativa de design viável para aplicações de alerta precoce educacional em contextos onde o processamento e a conectividade são limitados, enfatizando a utilidade preditiva, a execução local, o controlo institucional e a supervisão humana.

Keywords
Artificial intelligence in educationFrugal AIEducational early-warning systemsStudent dropout predictionEdge AIDigital sovereignty

References 30

  1. Al-Zawqari, A. M. M., Peumans, D., & Vandersteen, G. (2024). Latent space bias mitigation for predicting at-risk students. Computers and Education: Artificial Intelligence, 7, 100300. https://doi.org/10.1016/j.caeai.2024.100300
  2. Alfredo, R., Echeverría, V., Jin, Y., Yan, L., Swiecki, Z., Gašević, D., & Martinez-Maldonado, R. (2024). Human-centred learning analytics and AI in education: A systematic literature review. Computers and Education: Artificial Intelligence, 6, 100215. https://doi.org/10.1016/j.caeai.2024.100215
  3. Bañeres, D., Rodríguez-González, M. E., Guerrero-Roldán, A.-E., & Cortadas, P. (2023). An early warning system to identify and intervene online dropout learners. International Journal of Educational Technology in Higher Education, 20, 3. https://doi.org/10.1186/s41239-022-00371-5
  4. Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5–32. https://doi.org/10.1023/A:1010933404324
  5. Christou, V., Tsoulos, I., Loupas, V., Tzallas, A. T., Gogos, C., Karvelis, P. S., Antoniadis, N., Glavas, E., & Giannakeas, N. (2023). Performance and early drop prediction for higher education students using machine learning. Expert Systems with Applications, 225, 120079. https://doi.org/10.1016/j.eswa.2023.120079
  6. Fratini, S., Hine, E., Novelli, C., Roberts, H., & Floridi, L. (2024). Digital sovereignty: A descriptive analysis and a critical evaluation of existing models. Digital Society, 3, 59. https://doi.org/10.1007/s44206-024-00146-7
  7. Gunasekara, S., & Saarela, M. (2026). Trade-offs between fairness and performance in educational AI: Analyzing post-processing bias mitigation on the OULAD. Information and Software Technology, 189, 107933. https://doi.org/10.1016/j.infsof.2025.107933
  8. Herrmann, L., & Weigert, J. (2024). AI-based prediction of academic success: Support for many, disadvantage for some? Computers and Education: Artificial Intelligence, 7, 100303. https://doi.org/10.1016/j.caeai.2024.100303
  9. Hevner, A. R., March, S. T., Park, J., & Ram, S. (2004). Design science in information systems research. MIS Quarterly, 28(1), 75–105. https://doi.org/10.2307/25148625
  10. Idowu, J. A., Koshiyama, A. S., & Treleaven, P. (2024). Investigating algorithmic bias in student progress monitoring. Computers and Education: Artificial Intelligence, 6, 100267. https://doi.org/10.1016/j.caeai.2024.100267
  11. Kuzilek, J., Hlosta, M., & Zdrahal, Z. (2017). Open University Learning Analytics dataset. Scientific Data, 4, 170171. https://doi.org/10.1038/sdata.2017.171
  12. Lopes, A., Santos, F. P., Oliveira, D., Schiezaro, M., & Pedrini, H. (2024). Computer vision model compression techniques for embedded systems: A survey. Computers & Graphics, 123, 104015. https://doi.org/10.1016/j.cag.2024.104015
  13. Marcolino, M. R., Porto, T. R., Primo, T. T., Targino, R., Ramos, V., Queiroga, E. M., Munoz, R., & Cechinel, C. (2025). Student dropout prediction through machine learning optimization: Insights from Moodle log data. Scientific Reports, 15, 9840. https://doi.org/10.1038/s41598-025-93918-1
  14. Miao, F., Holmes, W., Huang, R., & Zhang, H. (2021). AI and education: Guidance for policymakers. UNESCO. https://doi.org/10.54675/PCGD1037
  15. Paolucci, C., Vancini, S., Bex, R. T., Cavanaugh, C., Salama, C., & de Araujo, Z. (2024). A review of learning analytics opportunities and challenges for K-12 education. Heliyon, 10(4), e25767. https://doi.org/10.1016/j.heliyon.2024.e25767
  16. Peffers, K., Tuunanen, T., Rothenberger, M. A., & Chatterjee, S. (2007). A design science research methodology for information systems research. Journal of Management Information Systems, 24(3), 45–77. https://doi.org/10.2753/MIS0742-1222240302
  17. Ray, P. P. (2023). A review on TinyML: State-of-the-art, challenges and future directions. Journal of King Saud University—Computer and Information Sciences, 35(1), 312–330. https://doi.org/10.1016/j.jksuci.2021.11.012
  18. Satyanarayanan, M. (2017). The emergence of edge computing. Computer, 50(1), 30–39. https://doi.org/10.1109/MC.2017.9
  19. Shi, W., Cao, J., Zhang, Q., Li, Y., & Xu, L. (2016). Edge computing: Vision and challenges. IEEE Internet of Things Journal, 3(5), 637–646. https://doi.org/10.1109/JIOT.2016.2579198
  20. Sun, K., Wang, X., Miao, X., & Zhao, Q. (2025). A review of AI edge devices and lightweight CNN and LLM deployment. Neurocomputing, 614, 128791. https://doi.org/10.1016/j.neucom.2024.128791
  21. UNESCO. (2021). Recommendation on the ethics of artificial intelligence. UNESCO.
  22. UNESCO. (2023). Guidance for generative AI in education and research. UNESCO. https://doi.org/10.54675/EVFR4258
  23. Vaarma, M., & Li, H. (2024). Predicting student dropouts with machine learning: An empirical study in Finnish higher education. Technology in Society, 76, 102474. https://doi.org/10.1016/j.techsoc.2024.102474
  24. Warden, P., & Situnayake, D. (2019). TinyML: Machine learning with TensorFlow Lite on Arduino and ultra-low-power microcontrollers. O’Reilly Media.
  25. Williamson, B. (2021). Making markets through EdTech platforms: Commercializing educational data and AI. Critical Studies in Education, 62(3), 321–337. https://doi.org/10.1080/17508487.2020.1856152
  26. Chung, J. Y., & Lee, S. (2019). Dropout early warning systems for high school students using machine learning. Children and Youth Services Review, 96, 346–353. https://doi.org/10.1016/j.childyouth.2018.11.030
  27. Arnold, K. E., & Pistilli, M. D. (2012). Course Signals: Finding everyday analytics for student success. Proceedings of the 2nd International Conference on Learning Analytics and Knowledge, 267–270. https://doi.org/10.1145/2330601.2330667
  28. Hastie, T., Tibshirani, R., & Friedman, J. (2009). The elements of statistical learning (2nd ed.). Springer. https://doi.org/10.1007/978-0-387-84858-7
  29. Luckin, R., Holmes, W., Griffiths, M., & Forcier, L. B. (2016). Intelligence unleashed: An argument for AI in education. Pearson.
  30. Selwyn, N. (2019). Should robots replace teachers? AI and the future of education. Polity Press.

Cite this article

BibTeXRIS

Zotero and Mendeley: open the RIS file, or use the Zotero Connector on this page (it reads the article metadata).

Share

Related articles