Scholarly Journals Accepted 120 Fake Research Papers Generated by Computer Program
More than 120 computer-generated "gibberish" research papers are being removed from the archives of scientific journal publishers Springer and the Institute of Electrical and Electronic Engineers (IEEE) after a French computer scientist determined the papers were fakes.
The bogus research papers, it turns out, were created by an automated word generation program that can string random, seemingly sophisticated words together in plausible English syntax.
Scientific papers, especially those dealing with computer science and mathematics, as these fake papers were, feature reams of sophisticated jargon. Even legitimate papers can seem like gibberish to an unfamiliar reader.
Among the papers' titles were computer-generated gems such as "Application and Research of Smalltalk Harnessing Based on Game-Theoretic Symmetries"; "An Evaluation of E-Business with Fin"; and "Simulating Flip-Flop Gates Using Peer-to-Peer Methodologies," according to a Fox News report.
The computer scientist who unveiled the fake papers, Cyril Labbé of Joseph Fourier University in France, told Fox News that there is high pressure on research scientists to publish and to do so frequently, which creates an environment where publishing fake research can be incentivized.
"They all should have been evaluated by a peer-review process. I've no explanation for them being here. I guess each of them needs an investigation," Labbé told Fox News.
The Nature News blog reported that most of the these gibberish papers were submitted to research conferences in China and that some of the authors listed on the papers did not appear to be aware their name was on the research.
Labbé and his colleagues have created a piece of software called SCIgen which randomly combines strings of words to produce fake research papers. They created the software in 2005 to prove how easy it is to fake research, the Nature blog reported. SCIgen is free to download and use, but it is unclear how many people have done so and for what purposes they have used to software. They have also build a website that can detect whether a paper has been produced usign SCIgen.
In an unrelated but similar event, a 2013 study in the journal Science found that some online journals will publish bad or fictitious research for a fee. US journalist John Bohannon submitted a deliberately faked research article to more than 300 online scientific journals, more than half of them accepted the fake research for publication.