Search

The purpose of this study is to develop a text clustering–based analysis of COVID-19 research articles. Owing to the proliferation of published COVID-19 research articles, researchers need a method for reducing the number of articles they have to search through to find material relevant to their expertise. The study analyzes 83,264 abstracts from research articles related to COVID-19. The textual data are analysed using singular value decomposition (SVD) and the expectation–maximisation (EM) algorithm. Results suggest that text clustering can both reveal hidden research themes in the published literature related to COVID-19, and reduce the number of articles that researchers need to search through to find material relevant to their field of interest. [ABSTRACT FROM AUTHOR]

Over the past two decades, databases and the tools to access them in a simple manner have become increasingly available, allowing historical and modern-day topics to be merged and studied. Throughout the recent COVID-19 pandemic, for example, many researchers have reflected on whether any lessons learned from the Spanish flu pandemic of 1918 could have been helpful in the present pandemic. Most studies using text-mining applications rarely use full-text journal articles. This article provides a methodology used to develop a full-text journal article corpus using the R fulltext package. Using the proposed methodology, 2743 full-text journal articles were obtained. The aim of this article is to provide a methodology and supplementary codes for researchers to use the R fulltext package to curate a full-text journal corpus. [ABSTRACT FROM AUTHOR]