Research Trends in Artificial Intelligence for Learning Analytics: A Bibliometric Mapping

Artificial Intelligence (AI) has progressively been integrated into learning analytics, thereby enhancing data-driven decision-making processes within educational settings. Despite the rapid growth of this interdisciplinary field, a comprehensive understanding of its publication productivity, citation impact, and thematic evolution remains limited. To address this gap, this study presents a Scopus-based bibliometric mapping that integrates productivity indicators, citation impact analysis, and longitudinal thematic clustering to provide a structural overview of AI-driven learning analytics research. This study aims to map the research landscape of AI in learning analytics through a bibliometric analysis. Bibliographic data were sourced from the Scopus database, spanning publications from 2010 to 2025. A comprehensive analysis of 1,401 documents was conducted using VOSviewer to examine publication trends, research productivity and impact across countries, authors, and publications, as well as dominant research themes. The results show a significant increase in publications over the past few years, suggesting growing scholarly interest in AI-driven learning analytics. The United States, Australia, and China emerged as the most productive and influential countries, while a small group of highly cited authors and journals also demonstrated high research impact based on citation indicators. Keyword co-occurrence analysis revealed significant research themes, encompassing AI-driven educational analytics, predictive analytics modelling, adaptive learning systems, generative AI-driven learning analytics, immersive technologies, multimodal analytics, and natural language processing-based applications. In conclusion, this study provides a systematic mapping of research trends and the intellectual structure of AI-driven learning analytics. The findings offer strategic insights for researchers, educators, and policymakers by highlighting emerging methodological innovations, expanding application domains, and opportunities for interdisciplinary collaboration.