Research Dataset

This dataset supports the bibliometric analysis of ChatGPT applications in educational research, covering the period 2023.01–2025.12 (retrieval date: 2026.01.15).

1. Data Source: Raw records were retrieved from Web of Science Core Collection with the search strategy: ("ChatGPT" OR "Large Language Model") AND ("education" OR "pedagogy" OR "learning design")，limited to Article, Review (excluding conference abstracts and editorials).
2. Data Content: It includes 626 final literature entries (after duplicate removal, title/abstract screening, and full-text review) with fields: author names, affiliations, publication year, journal, keywords (author-supplied + indexed), citation count, and reference list. Processed data files (e.g., keyword co-occurrence matrix for 9 clusters，author collaboration network data) used in CiteSpace (v6.2.R4) and VOSviewer (v1.6.19) visualizations are also included.
3. Data Usage: This dataset enables replication of the study’s findings (research output trends, international collaboration patterns, thematic hotspots of ChatGPT in education) and can be reused for extended bibliometric or systematic review studies in this field.
4. Notes: Data screening followed PRISMA 2020 guidelines; visualization parameters (e.g., CiteSpace time slicing: 1 year, node type: Keyword) are consistent with the paper’s methodology section.