Abstract

This study aimed to analyze the evaluation indicators and research topics related to university libraries and suggested directions for improving the current evaluation framework. To achieve this, the study compared the results of topic modeling on university library-related research articles with existing evaluation indicators, and examined how these indicators were reflected in actual academic discourse. A total of 626 articles published between 2001 and 2024 in four major journals in the field of library and information science were collected, and the Latent Dirichlet Allocation (LDA) topic modeling technique was applied to extract key topics. As a result, 17 major topics were identified, including services, librarians, electronic resources, information literacy education, evaluation standards, and technology adoption. Of these, 14 topics aligned with the current evaluation indicators, while two topics—“data” and “technology adoption”—did not correspond to any existing indicators. Conversely, certain evaluation indicators such as education, budget, specialization, and library status had no corresponding research topics. These findings offered meaningful implications for the improvement of evaluation indicators and suggested future research directions in the field of university libraries