Analyzing the evolution of machine learning integration in educational research: a bibliometric perspective

Abstract Machine learning, a subset of artificial intelligence, has experienced rapid advancements and applications across various domains. In education, its integration holds great potential to revolutionize teaching, learning, and educational outcomes. Despite the growing interest, there needs to be more comprehensive bibliometric analyses that track the trajectory of machine learning’s integration into educational research. This study addresses this gap by providing a nuanced perspective derived from bibliometric insights. Using a dataset from 1986 to 2022, consisting of 449 documents from 145 sources retrieved from the Web of Science (WoS), the research employs network analysis to unveil collaborative clusters and identify influential authors. A temporal analysis of annual research output sheds light on evolving trends, while a thematic content analysis explores prevalent research themes through keyword frequency. The findings reveal that co-authorship network analysis exposes distinct clusters and influential figures shaping the landscape of machine learning in educational research. Scientific production over time reveals a significant surge in research output, indicating the field’s maturation. The co-occurrence analysis emphasizes a collective focus on student-centric outcomes and technology integration, with terms like “online” and “analytics” prevailing. This study provides a nuanced understanding of the collaborative and thematic fabric characterizing machine learning in educational research. The implications derived from the findings guide strategic collaborations, emphasizing the importance of cross-disciplinary engagement. Recommendations include investing in technological infrastructure and prioritizing student-centric research. The study contributes foundational insights to inform future endeavors in this ever-evolving field.