Exploring the Landscape of ChatGPT Research: A Bibliometric Study Using Scopus Database

Abstract

The analysis used in this paper shows that ChatGPT research is racing through its early life cycle, taking information from the Scopus database over the past several years and providing a holistic landscape of responses to each genre, respectively. By carefully extracting and analyzing 10,083 entries, this study illuminates which documents, authors, institutions, and countries are driving the research on what ChatGPT methods. The results from the co-authorship analysis suggest dominant patterns of collaboration, representing a few clusters of productive authors and contributing institutions, and show that countries such as the USA, China, India, and the UK contribute to both a higher number of documents counts and a higher citation impact. The main topics derived from our keyword co-occurrence analysis are “natural language processing,” “machine learning,” and the broader topic of artificial intelligence, plus emerging themes around ethics and applications. Bibliographic coupling shows how tightly knit research plots are and provides a clue about the relevance of leading journals as well as worldwide cooperation in the scheme. The results of this study will inform future directions for research, development, and policy to be effectively taken forward in a wide range of AI technology industries within the scholarly ecosystem of ChatGPT.

Ähnliche Arbeiten