Thematic shifts in education research as a mirror to Kazakhstan's development: a bibliometric and natural language processing analysis (1992–2024)

Purpose This study examines the evolution of education research in Kazakhstan from 1992 to 2024, a period marked by profound transformation, aiming to assess how scholarly output aligns with national education policy and broader developmental priorities. Design/methodology/approach A mixed bibliometric and natural language processing (NLP) approach was employed to analyze 956 English-language publications indexed in the Web of Science. The study combined quantitative analyses of publication trends, authorship and institutional collaboration with NLP techniques, including keyword co-occurrence, topic modeling and sentiment analysis, framed by a review of Kazakhstan’s education policy phases: post-Soviet foundational reforms and an era of global alignment. Findings Results reveal a substantial surge in research output, particularly after 2020, driven by a core set of highly productive authors and institutions. Thematic analyses indicate partial alignment with policy priorities, with multilingual education, digitalization and pedagogical innovation prominent, while inclusive education and social justice appear less substantively addressed. Collaboration networks display a core-periphery structure, with Nazarbayev University acting as the primary international and national hub. Originality/value This study offers the first comprehensive, longitudinal mapping of Kazakhstani education research, integrating bibliometric and NLP methods with policy analysis. It highlights both the rapid quantitative growth and the structural and thematic limitations of the field, providing a foundation for policy and institutional strategies aimed at fostering more diverse, critical and sustainable research.