The text mining approach to investigate active areas of library and information science and discover emerging topics

Abstract

Purpose This study aims to adopt text mining to discover emerging topics in librarianship and information science research in the last decade. Based on the number of citations obtained during the previous 10 years, the authors selected emerging topics in this study and evaluated the strength of their presence. Additionally, the authors determined if the trend was substantial over time and identified the active topics in library and information science (LIS) through the past 10 years. Design/methodology/approach All library and medical information studies were retrieved by the WC = “Information Science & Library Science” tag in the Web of Science. Python programming was used for data analysis. The topics were identified by combining the unsupervised deep learning algorithms TOP2VEC and the term frequency-inverse document frequency and also the Mann–Kendall trend test is used to determine whether the trend was significant over time. Findings Following text mining, the total data from 2012 to 2021 was 63,712. Eleven main topics were also extracted: academic education of LIS, acquisition and collection development, publishing articles, cataloging and classification, journalism, knowledge management, infometrics, social media, university ranking, information and communication technologies and information storage and retrieval. Knowledge management has experienced the greatest growth over the past 10 years. Originality/value This analysis reveals which fields are prioritized and which are neglected by the LIS. The findings of this study can help researchers discover newer topics, focus on less-seen subjects and prevent repetitive research in one area.