Libraries at University of Nebraska-Lincoln
Abstract
Biomedical literature is increasing day by day. The present scenario shows that the volume of literature regarding “coronavirus” has expanded at a high rate. In this study, text mining technique has been employed to discover something new from the published literature. The main objectives of this study are to show the growth of literature (Jan-Jun, 2020), extract document section, identify latent topics, find the most frequent word, represent the bag of words, and the hierarchical clustering. We have collected 16500 documents from PubMed. This study finds most number of documents (11499) belong to May and June. We explore “betacoronavirus” as the leading document section (3837); “covid” (29890) as the most frequent word in the abstracts; and positive-negative weights of topics. Further, we measure the term frequency (TF) of a document title in the bag of words model. Then we compute a hierarchical clustering of document titles. It reveals that the lowest distance the selected cluster (C133) is 0.30. We also have made a discussion over future prospects and mentioned that this paper can be useful to researchers and library professionals for knowledge management.