The Post-ChatGPT Research Landscape of Generative Artificial Intelligence in Higher Education (2023–2026): A Bibliometric Analysis - Figure 4

The co-occurrence map of terms identified four major clusters, outlining both the educational core of research on GenAI in higher education and several related technical and disciplinary areas. Figure 4 illustrates the thematic structure of the field through a keyword co-occurrence analysis (minimum threshold of 150 terms), highlighting a densely interconnected network organised around several dominant conceptual hubs. The terms “artificial intelligence” and “large language models” occupy central positions both in terms of node size (high frequency) and the number of connections, confirming the role of technological infrastructure as the central pivot of the post-ChatGPT literature.