摘要：

Abstract:

【Purpose/significance】The purpose of this study is to explore the distribution characteristics of library-themed texts in the

AI environment, aiming to provide practical references and theoretical support for the intelligent transformation and service upgrade of

libraries.【Method/process】This study is based on the core collection of the Web of Science database, using the BERTopic model to

identify topics and extract knowledge structures from the retrieved related literature, to gain insight into the current research status and

future development trends of AI in the library field

.

【Result/conclusion】The research results show that the main research topics of li⁃

braries in the AI environment can be divided into five categories: library intelligence, collection resource development, academic re⁃

search support, information resource construction, and chatbots & intelligent agents. Finally, combined with topic content analysis, the

opportunities and challenges of libraries in the AI environment are discussed in depth.【Innovation/limitation】This study leverages the

cutting-edge natural language processing technology, BERTopic, to extract knowledge from topic-specific texts in the library domain

within an AI context. Future investigations will expand to include a wider variety of data sources, aiming for more holistic and meticu⁃

lous exploration.