A Data-Driven Exploration of Emerging Library Trends Using Topic Modeling

Abstract

The recent digital revolution of Library and Information Science (LIS) has transformed bibliographic data into huge datasets, which have no organised structure and can only be explored through sophisticated analytical systems. This paper uses Latent Dirichlet Allocation (LDA) topic modeling of 279 publications (1974-2024) of Scopus indexed publications on emerging library trends. After undergoing intensive text preprocessing on RStudio, there came up five overarching themes, namely, digital preservation, bibliometric analysis, digital libraries and services, online learning platforms, and emerging LIS trends. Findings reveal that digital preservation (approximately 30 % of topic coverage) and equal coverage of bibliometrics, digital services, and online learning (approximately 20 % each) were the areas of strong research, and the emergent trends include approximately 10 % of coverage. These results emphasise the growing adoption of digital technologies, data analysis and preservation policies in LIS research. The research paper provides viable suggestions that libraries, especially in the developing world can implement to pursue data-driven planning, enhance digital preservation policies, and train employees to be proficient in computational practices. Projecting these themes on current global and Indian LIS trends, the study offers comparative evidence and practice based suggestions on how to empower digital library frameworks and expert abilities.