Exploring Scholarly Journal Content: Topic Modeling Analysis Using LDA Algorithm

This book section presents a topic modeling analysis employing the Latent Dirichlet Allocation (LDA) algorithm to examine the content of the Journal of Information Science. The analysis adopts a data-driven approach to comprehend the diverse topics and themes present in the journal's content, providing valuable insights for journal authors, editors, and readers. The study further incorporates a topic modeling-based bibliometric analysis of published articles, offering a comprehensive overview of the journal. The bibliometric analysis results reveal significant inferences about the journal's leading countries, affiliations, and authors based on the publications and citation trends. The topic modeling analysis classifies articles under 18 distinct topics, with “Decision analysis” and “Library science” emerging as the two most frequently explored subjects. Additionally, acceleration results for each topic are examined both independently and comparatively with other topics. The findings indicate that “Social networks” exhibited the highest acceleration compared to other topics, while “Data mining” achieved the highest acceleration. In conclusion, this paper demonstrates the effectiveness of LDA-based topic modeling as a robust tool for analyzing journal content. Such data-driven analyses can effectively support journal management, enhance content strategies, and foster increased reader engagement, providing a fundamental basis for future research endeavors.