Artificial intelligence and machine learning have rapidly expanded their presence in educational research, but these labels encompass very different applications, techniques, and problems. The systematic review by Wiston Forero-Corba and Francisca Negre Bennasar analyses 55 studies published between 2021 and February 2023 in Web of Science and Scopus, covering 38 countries and educational levels ranging from primary school to university.
The study identifies 33 artificial intelligence and machine learning techniques and examines what they were being used for: from predicting academic performance and student dropout to academic guidance, robotics, recommender systems, the teaching of artificial intelligence itself, and support for specific groups of students. Rather than revealing a single dominant trend, the review portrays a diverse field in which intelligent technologies are beginning to play a role in both learning and educational management.
One of the most significant findings is that this research is no longer concentrated primarily in higher education: 74.6% of the studies analysed were conducted in primary or secondary education. There is also a clear predominance of supervised learning and predictive applications, particularly those related to academic performance, for which techniques such as Random Forest are frequently used. Yet the picture is broader: the review includes experiences involving computational thinking, AI literacy, robotics, academic guidance, inclusion, and personalised learning.
This diversity highlights both the possibilities offered by these technologies and the need for teachers who are able to understand them and integrate them in pedagogically appropriate ways. The authors therefore emphasise the importance of strengthening teachers’ digital competence and also point out that the intensive use of data raises questions about data quality, potential biases, and responsible handling.
The study thus provides a broad snapshot of research on artificial intelligence and machine learning in education at a time of rapid growth in the field. That snapshot also has limitations: the review considers only English-language articles indexed in WoS and Scopus, reveals an uneven geographical distribution, and identifies a still limited presence of research on diversity, disability, and special educational needs. Nevertheless, its findings help shift the discussion away from an overly general question (what can artificial intelligence do in education?) towards more specific ones: what educational problems can it help address, which techniques are appropriate for each purpose, and what preparation do teachers and institutions need to use them effectively and responsibly?
The expansion of these technologies, the study suggests, does not in itself turn AI into educational innovation; its value depends on how and why it is incorporated into teaching, learning, and educational management processes.
---
How to Cite: Forero-Corba, W., & Negre Bennasar, F. (2024). Techniques and applications of Machine Learning and Artificial Intelligence in education: a systematic review. RIED-Revista Iberoamericana de Educación a Distancia, 27(1), 209–253. https://doi.org/10.5944/ried.27.1.37491
