Large Language Models (LLMs): A systematic study in Administration and Business.

Purpose: With the advancement of the use of LLMs, there is a growing need to understand the current research scenario and potential trends and gaps in this field of knowledge. Therefore, bibliometric analysis was used with the aim of analyzing scientific production involving applications of LLMs in Administration and Business. Originality/value: This study analyzes the characteristics of academic production involving LLMs, Administration, and Business and provides potential insights for researchers and professionals in the field. Design/methodology/approach: To achieve the objectives of this work, bibliometrics and systematic mapping were conducted from 2000 to 2024 to answer the following questions: What is the state of the art of academic production involving LLMs in Administration and Business? What is the state of the art of empirical studies involving LLMs in Administration and Business? What is the focus of LLM applications in Administration and Business? Findings: Most articles involve computational modeling and empirical analyses and refer to validating existing technologies, methods, or tools. The research was classified according to 6 categories regarding the application objectives of LLMs: Tracking, Recognition, Extraction, Modeling, Summarization, and Classification. The systematic map analysis indicates that, despite advances in the use and application of LLMs, some challenges persist and represent possibilities for future research. Issues involving data ethics and privacy and the management of research biases involving natural language processing are prominent challenges. [ABSTRACT FROM AUTHOR]