A veritable crisis has come up for a number of respected science journal publishers. The discovery of over a hundred papers filled with computer-generated gibberish has publishers scrambling to clean up what could be an ugly mess.
Springer and the Institute of Electrical and Electronics Engineers (IEEE), both respected publishers, are now taking down over 120 papers from their archives after it was discovered that the papers contained nothing by computer-generated blather.
The problem was first discovered by a French researcher two years ago. Cyril Labbé, a computer scientists from the Joseph Fournier University in France, has spent the past couple of years tracking down these fake papers. What's surprising is that Labbé found many of the offending papers published in over 30 conference proceedings since 2008. Labbé's research also showed that over a hundred of the papers were published in the IEEE while 16 papers were found in previous publications of Springer.
Labbé has confirmed that the papers were created using a program called SCIgen, which was developed by graduate students from the Massachusetts Institute of Technology (MIT) back in 2005. The program has the capability to generate nonsense science papers. The creators of the program said that they made SCIgen to prove that fake papers could make it past respected publishers and conferences.
"SCIgen is a program that generates random Computer Science research papers, including graphs, figures, and citations," said the program creators on the SCIgen webpage. "It uses a hand-written context-free grammar to form all elements of the papers. Our aim here is to maximize amusement, rather than coherence."
Labbé has been busy developing a method of automatically tracking and finding these computer-generated papers. However, the computer scientist has stated that he is not aware of the motives behind the fake submissions. Most of the conferences involved where found in China and many of the fake papers were also associated with Chinese affiliations.
"The papers are quite easy to spot," says Labbé.
Labbé has already contacted many of the affected parties privately regarding the computer-generated papers. It is currently unknown how many of the affected authors were aware of the situation. However, one author has come forward saying that he intentionally submitted a fake paper just to test out a certain conference.
"I wasn't aware of the scale of the problem, but I knew it definitely happens. We do get occasional e-mails from good citizens letting us know where SCIgen papers show up," says SCIgen co-creator Jeremy Stribling.
