R code to read and analyse data to examine the use of acronyms in published papers over time. The analysis examines titles and abstracts published on PubMed up until 2019. Our definition of acronym includes initialisms and abbreviations.
The folder animation contains animations of the top ten acronyms per year over time in titles and abstracts.
The folder data contains the following data on acronyms and the meta-data on papers:
titles[x].rdsmeta-data on the 24,873,372 included titles in rds formattitles_sample.txta random sample of 1,000 included titles in tab-delimited formatabstracts[x].rdsmeta-data on the 18,249,091 included titles in rds formatabstracts_sample.txta random sample of 1,000 included abstracts fromabstracts.RDSin tab-delimited formatacronyms[x].rdsthe 139,959,947 acronymsacronyms_sample.txta random sample of acronyms from 1,000 papers in tab-delimited format The data are very large and hence have been split into multiple files. For the tab-delimited files I've given a random sample as an easily accessible taster of the data.
The data were sourced directly from PubMed in XML format (available here) hosted by the National Library of Medicine. The data here do not reflect the most current/accurate data available from the National Library of Medicine. The data were downloaded between 14 to 22 April 2020.
The variables in title[x].rds, titles_sample.txt, abstracts[x].rds and abstracts_sample.txt are:
pmidPubMed ID numberdatedate published on PubMedtypearticle type, e.g., "Journal Article" or "Editorial"jabbrvjournal abbreviation, e.g., "Biochem Med"n.authorsnumber of authorsn.wordsnumber of words in the title or abstract
The variables in acronyms_sample.txt and acronyms[x].rds are:
pmidPubMed ID numberacronymsthe acronym (e.g., "HIV")ncharthe number of characters in the acronymsource'Title' or 'Abstract'
The acronyms used above are:
- RDS = R data source???
- XML = Extensible markup language