Using R to develop a corpus of full-text journal articles

Abstract

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.