Sign In

Communications of the ACM

ACM TechNews

'Papermill Alarm' Software Flags Potentially Fake Papers

View as: Print Mobile App Share:
A stack of research papers.

Anna Abalkina, an economist at the Free University of Berlin who studies paper mills, says the scientific community will benefit from automated checks that can detect potentially bogus papers.

Credit: Raimund Koch/Getty

Adam Day at U.K.-based data services company Clear Skies developed a software tool that highlights potentially bogus scientific papers.

The Papermill Alarm uses a deep learning algorithm to analyze documents' titles and abstracts for language similar to that contained in the titles and abstracts of fake articles.

The tool color-flags papers red for high similarity to false content, orange for some similarity, and green for no similarity.

Day said he processed all titles listed in the PubMed citation database, and the Papermill Alarm flagged 1% of currently listed papers as having text similar to that of articles produced by paper mills, companies, or individuals that fabricate scientific manuscripts to order.

From Nature
View Full Article


Abstracts Copyright © 2022 SmithBucklin, Washington, DC, USA


No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account