Science publisher fooled by gibberish papers

Feb 27, 2014 by Richard Ingham And Laurent Banguet

Publisher of science journals Springer said Thursday it would scrap 16 papers from its archives after they were revealed to be computer-generated gibberish.

The fake papers had been submitted to conferences on computer science and engineering whose proceedings were published in specialised, subscription-only publications, Springer said.

"We are in the process of taking down the papers as quickly as possible," the German-based publisher said in a statement.

"This means that they will be removed, not retracted, since they are all nonsense."

Springer added: "We are looking into our procedures to find the weakness that could allow something like this to happen, and we will adapt our processes to ensure that it does not happen again."

The embarrassing lapse was exposed by French computer scientist Cyril Labbe of the Joseph Fourier University in Grenoble.

He also spotted more than 100 other "nonsense" papers unwittingly published by the New York-based Institute of Electrical and Electronic Engineers (IEEE), the journal Nature reported.

In a statement to AFP, the institute said it had been advised "there might have been some conference papers published in our IEEE Xplore digital library that did not meet our quality standards."

"We took immediate action to remove those papers, and also refined our processes to prevent papers not meeting our standards from being published in the future," it said. The statement gave no further details.

Labbe, 41, has been exploring how to detect fake papers written with a programme called SCIgen.

At the press of a button, the programme cranks out impressive-looking "studies" stuffed with randomly-selected computer and engineering terms.

Instant goobledook

Here is an example: "Constant-time technology and access points have garnered great interest from both futurists and physicists in the last several years. After years of extensive research into superpages, we confirm the appropriate unification of 128-bit architectures and checksums."

This "paper" comes complete with fake graphs and citations—essential features in scientific publishing—that in SCIgen's case includes recent references to famous scientists who died decades or centuries ago.

The programme was devised in 2005 by researchers at the Massachusetts Institute of Technology.

They used it to concoct meaningless papers that were accepted by conferences. The researchers later revealed the hoax to expose flaws in safeguards.

SCIgen is freely available online, at pdos.csail.mit.edu/scigen/

Labbe told AFP he spotted the frauds by searching for telltale SCIgen vocabulary.

In 2010, he used SCIgen to create 102 bogus papers by a fictitious scientist and added these to the Google Scholar database, an index of science prestige.

For a time, "Ike Antkare" ranked 21st on the database's list of most-cited scientists in the world—higher than Einstein, who ranked a lowly 36th.

The fake papers detected by Labbe were submitted to conferences between 2008 and 2013. They were uncovered through research he published in 2012 in Scientometrics—by coincidence, also a Springer journal.

In some cases, he said, a paper's introduction or conclusion were rewritten by a human to appear more authentic at first glance—a veneer presumably aimed at fooling superficial scrutiny.

'Peer review' under pressure

Labbe said the fraud struck at the credibility of peer-reviewed systems in which scientific claims are meant to be assessed by independent experts for soundness.

"There are several possible explanations" for the fakes, he said.

"One is that people are just testing the system, but if that's the case, they should reveal who they are and they haven't done so," said Labbe. "Another is that the papers are a deliberate fraud to make money."

Springer said scientific publishing, like other fields, "is not immune to fraud and mistakes".

"The peer-review system is the best system we have so far and this incident will lead to additional measures on the part of Springer to strengthen it."

Also on Thursday, South Korea's Supreme Court upheld an 18-month suspended jail term against Hwang Woo-Suk, accused of embezzlement and abuse of ethics in one of the most notorious frauds in science publishing.

Hwang shot to fame in 2004 when he published papers in the prestigious US journal Science claiming to have created the first stem-cell lines from a cloned human embryo.

The claims raised hopes of new treatments for diseases like cancer, diabetes and Parkinson's, and Hwang and his team were showered with money and national honours.

His findings were later found to have been faked. No stem cells had been produced.

Explore further: How the 'Matthew Effect' helps some scientific papers gain popularity

add to favorites email to friend print save as pdf

Related Stories

Data are lost to science at 'astonishing rate'

Dec 19, 2013

New evidence reported in the journal Current Biology on December 19 confirms long-held fears about the fate of scientific data. Careful evaluation of more than 500 randomly selected studies found that the original data b ...

Recommended for you

Research band at Karolinska tuck Dylan gems into papers

22 hours ago

(Phys.org) —A 17-year old bet among scientists at the Karolinska Institute has been a wager that whoever wrote the most articles with Dylan quotes before they retired would get a free lunch. Results included ...

A simulation game to help people prep for court

Sep 25, 2014

Preparing for court and appearing before a judge can be a daunting experience, particularly for people who are representing themselves because they can't afford a lawyer or simply don't know all the ropes ...

When finding 'nothing' means something

Sep 25, 2014

Scientists usually communicate their latest findings by publishing results as scientific papers in journals that are almost always accessible online (albeit often at a price), ensuring fast sharing of latest ...

User comments : 29

Adjust slider to filter visible comments by rank

Display comments: newest first

julianpenrod
1.4 / 5 (9) Feb 27, 2014
It's a good question what really constitutes "scientific" knowledge, these days. Apparently, "peer review" doesn't go much deeper than to make sure random words are spelled correctly. Hundreds, even thousands of other papers were found to describe phenomena which were non reproducible. Yet how much coverage in the "news" does that receive? Reams and reams of "results" passed off as "real". And dependent in their fraud on, among other things, the characteristic of blind, unthinking acceptance by dull witted myrmidons who hate God and look for any "excuse" to pretend He does not exist, or who look to parroting "science" claims to disguise their own rancid failure to achieve. And, if something goes unchallenged or claimed true but not examined by the entire population as a whole, like the lies about "fossils", "evolution", "relativity", the "safety" of vaccines, where is the force that will stop the lie from spreading?
Q-Star
4.4 / 5 (7) Feb 27, 2014
Never mind, carry on. I thought maybe Zeph had made it into print.
Mimath224
5 / 5 (2) Feb 27, 2014
People like Hwang Woo-Suk should be thrown into a ring and left to the people whose hopes he falsely rasied. I'm sure other fakes cause a similar problem in the areas that they publish.
dedereu
5 / 5 (3) Feb 27, 2014
Refereeing seriously a paper is a lot of work, which, for no errors needs, to reproduce the experiments, or the calculations, which can be made only by those working exactly on the same problem. Finding the right specialist expert is difficult to publishers sometimes.
Nearly no referee is working sufficiently, so that papers with no disturbing results are accepted, even fraud or plagiary.
But many real disturbing discoveries have been refused, even for future Nobel Prizes.
In conferences, often, papers are less refereed, so that crazy non sense papers can be accepted !!
antialias_physorg
4.2 / 5 (6) Feb 28, 2014
I wonder how many such papers are present in non-peer reviewed collections (arxiv?)

Peer review is certainly not infallible. But I have to - grudgingly - agree with the Springer official: it's the best system we currently have.
You always have to remember that a research paper is something no one else has done before. So the only person who really has an in-depth understanding at that point is the author (and maybe not even him/her if the paper is faulty).
Peer reviewers come from the same field, but not from the same specialty. They should generally be able to see whether a paper makes sense or not - but the specifics...?

It might be helpful if reviewers could have (moderated/anonymized) conversations about the content of the paper to make sure they understand it.
Rimino
Feb 28, 2014
This comment has been removed by a moderator.
Rimino
Feb 28, 2014
This comment has been removed by a moderator.
Captain Stumpy
4.3 / 5 (6) Feb 28, 2014
Also, the edditors are sniffing for crackpotism carefully with using of crackpot index. The articles which avoid it systematically will pass more freely just because of it.

@zeph
not so sure about this...
the 100+ papers pulled were from IEEE, which is where Electric Universe publishes its pseudoscience crap
probably why so many were pulled from their site
Most of my ideas are recognized immediately here just after single post, because they're unique

your stilted syntax also helps with recognition
TheGhostofOtto1923
2.8 / 5 (5) Feb 28, 2014
Soon enough computers will be writing genuine papers. They will also assume the task of peer review and will be far better at it than biased and unconcerned human peers.

And they will begin the task of sifting through our vast accumulation of knowledge and weeding out ALL gibberish.
antialias_physorg
4.1 / 5 (7) Feb 28, 2014
Most of my ideas are recognized immediately here just after single post, because they're unique

I wouldn't use the word 'unique'. I'd use the word 'special' (as in 'short bus special'). That kind of uniqueness is pretty easy to spot.
TheGhostofOtto1923
1.8 / 5 (6) Feb 28, 2014
Most of my ideas are recognized immediately here just after single post, because they're unique

I wouldn't use the word 'unique'. I'd use the word 'special' (as in 'short bus special'). That kind of uniqueness is pretty easy to spot.
So aa thinks it's ok to call somebody a retard if he's clever about it? How... revealing.

Soon enough computers will note crassness such as this and your bank account will be automatically debited and your credit report adjusted accordingly. And your post will be gang-rated appropriately.
Rimino
Feb 28, 2014
This comment has been removed by a moderator.
Whydening Gyre
3.7 / 5 (3) Feb 28, 2014
Soon enough computers will be writing genuine papers
For who? Such an articles would be good for computers only again. So that the http://phys.org/n...ans.html and we will just watch it helplessly, because we will not able to distinguish their intelligence from random noise - in the same way, like we cannot recognize the God in CMBR noise...;-) I just hope, such a research would be of the same practical relevance.

until we become one with them...:-)
Mimath224
3.7 / 5 (3) Feb 28, 2014
@Whydening Gyre Ha, do you mean 'resistance is futile' type relationship?
I realise that computers will become more sophisticated but wouldn't it have to be some kind of 'biocomp' that would think for itself? If a comp is mechanical what type of limit must be reached so G in G out is no longer applicable?
Whydening Gyre
not rated yet Feb 28, 2014
@Whydening Gyre Ha, do you mean 'resistance is futile' type relationship?
I realise that computers will become more sophisticated but wouldn't it have to be some kind of 'biocomp' that would think for itself? If a comp is mechanical what type of limit must be reached so G in G out is no longer applicable?

If it is possible, it is probable...
adam_russell_9615
4.8 / 5 (5) Mar 01, 2014
The journal is questioning whether the papers were a fraud, but if the journal is supposed to provide peer review and they were NOT, then isnt it the journal that is the fraud?
Bonia
Mar 01, 2014
This comment has been removed by a moderator.
Whydening Gyre
5 / 5 (1) Mar 01, 2014
@Whydening Gyre Ha, do you mean 'resistance is futile' type relationship?
I realise that computers will become more sophisticated but wouldn't it have to be some kind of 'biocomp' that would think for itself? If a comp is mechanical what type of limit must be reached so G in G out is no longer applicable?

Mimath. Not at all. given the correct algorithmic function, they can think and act as biological entity, even if not biologically constructed..\ That said, I believe we are using the wrong number base to correctly instigate this algorithmic function.
Mimath224
not rated yet Mar 01, 2014
Whydening Gyre are you thinking of 'mulitvalued logics' (modal logic)? That is P having a value of say 1, ½,0 (e.g.true, likely. not true) Or change the artimetic base from say 10 to 5 or other?
Whydening Gyre
not rated yet Mar 01, 2014
Whydening Gyre are you thinking of 'mulitvalued logics' (modal logic)? That is P having a value of say 1, ½,0 (e.g.true, likely. not true) Or change the artimetic base from say 10 to 5 or other?

6. So that thirds are possible. Either that or 60... Or even maybe 9.
cantdrive85
1 / 5 (4) Mar 01, 2014
not so sure about this...
the 100+ papers pulled were from IEEE, which is where Electric Universe publishes its pseudoscience crap
probably why so many were pulled from their site

Still lying?

Being that they pulled the unacceptable papers immediately and that Thornhill's, Scott's, et al are still there...

http://ieeexplore...rch=true

http://ieeexplore...rch=true

http://ieeexplore...rch=true

...would only indicate you're lying again.
Captain Stumpy
5 / 5 (2) Mar 02, 2014
Still lying?

@cant see/think
not lying, see: http://www.nature...-1.14763
I will quote (since you are too stupid to follow the link and read for yourself):
...and more than 100 were published by the Institute of Electrical and Electronic Engineers (IEEE), based in New York. Both publishers, which were privately informed by Labbé, say that they are now removing the papers.

[sic]
means that IEEE had hundreds pulled, like I said
my statement stand as perfectly factual
IEEE is where EU publishes because no reputable astrophysical journal will touch PSEUDOSCIENCE

and I guessed that was why so many were pulled from their site
given that we cannot READ the pulled papers, I would still say the same thing
any site willing to publish EU rubbish is not a reputable site

and NO I will not look at your links... I DO NOT READ garbage from sites that are not reputable

linking them only proves PSEUDOSCIENCE

not REAL SCIENCE
Captain Stumpy
5 / 5 (2) Mar 02, 2014
and I guessed

@cd
also, given that I gave a personal conjecture about thinking that the papers pulled were EU, I will state that it is based upon the following logic:
1- EU is a PSEUDOSCIENCE
2- EU supports the physically impossible and claims it as gospel for their astrophysical model
3- EU published papers on plasma physics and cosmology DO NOT CONTAIN ASTROPHYSICS and are related only by conjecture, or supposition due to faulty assessment of visible similarities and illogical extrapolation (see Moon craters/Saturn/Grand Canyon
4- ANY JOURNAL that would publish astrophysics papers without astrophysicists involved in it, or without an astrophysicist to proof it and assess its credibility, is not reputable
5- ANY JOURNAL that publishes KNOWN FALLACIES(that can be researched in a 2 minute internet search and using only reputable sites for resources) IS NOT A REPUTABLE JOURNAL

eag97a
5 / 5 (1) Mar 02, 2014
Isn't there some sort of a vetting process for all of these peer reviews done for papers being submitted for publication? These journals should at least require proof that panel defense has been done by the authors for the paper or at least there is indication that the authors for papers submitted has been personally vetted in some way. You would think that the editors will apply the scientific methodology in something as important as the peer review process.
cantdrive85
1 / 5 (4) Mar 02, 2014
and NO I will not look at your links... I DO NOT READ garbage from sites that are not reputable

I choose ignorance! CS
Captain Stumpy
5 / 5 (2) Mar 02, 2014
I choose ignorance! CS

@cant think
I make this special exception just for you and EU acolytes
and there is good reason
when I actually went to those links and tried to read those studies, I kept getting the same results... that you were pushing known pseudoscience

therefore, yes
I CHOOSE TO IGNORE THEM
and I choose to ignore them because it is simple...
IF THERE IS REPUTABLE SCIENCE, THEN YOU SHOULD BE ABLE TO FIND A REPUTABLE SOURCE TO SUPPORT IT, NOT SOME ELECTRICAL ENGINEERING JOURNAL THAT HAS NO RELATION OR IMPACT ON ASTROPHYSICS

simple, really

GIVEN that you HAVE YET to be able to do this
I can also then conclude
YOU ARE A PSEUDOSCIENCE TROLL

PERIOD
Bonia
Mar 02, 2014
This comment has been removed by a moderator.
Bonia
Mar 02, 2014
This comment has been removed by a moderator.
Captain Stumpy
5 / 5 (2) Mar 02, 2014
This attitude leads into pluralistic ignorance, which is as harmful from long term perspective, like the noncritical acceptation of ideas

@Zeph
you have a point about that
I just didnt like wasting time/resources/etc on something I knew was going to be a dud

I will concede that it is ignorant of me to ignore them without giving them the benefit of at least a cursory investigation/research

I will still argue:
using a site that is known for pushing/publishing pseudoscience is the same as linking to pictures of fairies climbing up the nose of a unicorn
IOW – irrelevant and without merit

IF there is real science AND
IF there is a legitimate point to be made
THEN
there WILL be studies from reputable journals that can support the statements
THEREFORE
I will still require proof to be from LEGITIMATE reputable sites
Captain Stumpy
5 / 5 (2) Mar 02, 2014
But we, as a laymen, we shouldn't constrain our interest to peer reviewed studies only, because we will face their skeptical bias soon. The judging of quality of scientific information is hard time consuming work - but if you avoid it and replace it with plain judging by source of this information, you should face the consequences

@Zeph
this really depends upon the circumstances IMHO
being INTERESTED is DIFFERENT than PROVING A POINT
IF we allow ANY source to be legitimate to PROVE a point, then we invite irrelevance into the argument, and we are no better than pseudoscientists
THEREFORE, like I said above, it is IMPERATIVE that the source of PROOF (empirical data) be a reputable source that is known for publications that are RELEVANT, LEGITIMATE and PEER REVIEWED...

but as far as INTEREST goes, I agree with you. There should be NO constraints/limitations and in this I usually have none

Maggnus
5 / 5 (4) Mar 02, 2014
@ Captain Stumpy - it's not actually "no I will not look at your links" it's "no I will not look at your links again!". It is that oft repeated tactic of the pseudoscientific - repeatedly post the same thing over and over in the vain hope it will somehow be more right (actually less wrong) than it was the first times it was posted.
Bonia
Mar 02, 2014
This comment has been removed by a moderator.
Mimath224
5 / 5 (2) Mar 02, 2014
@Captain Stumpy, was reading over the w/e about about a team of researchers from 3 univ published last month that teaching 'scientific reproducibility' has to be done at an early age. How I interpreted their ideas was that the scandals regarding bad papers and views was due to a lack of statistcal (math) training and quoted in particular, Biololgists. It must also be important to train people to follow basic rules and show all work. However, nothing was said about those who read and peer view or how to combat present problems. I don't think ststs training is the answer although it does help.
Captain Stumpy
5 / 5 (1) Mar 02, 2014
@Captain Stumpy, was reading over the w/e about about a team of researchers from 3 univ published last month that teaching 'scientific reproducibility' has to be done at an early age

@Mimath224
are you referring to the Science Mag article from january in the links below?

https://www.scien....summary

https://www.scien...229.full

if not... can you send me some links? I would like to read what your are talking about
sounds VERY INTERESTING

THANKS
Mimath224
5 / 5 (2) Mar 03, 2014
Captain Stumpy My pleasure
Duke University. "To teach scientific reproducibility, start young." ScienceDaily. ScienceDaily, 28 February 2014.
http://www.scienc...0138.htm
Captain Stumpy
5 / 5 (2) Mar 03, 2014
@Mimath224
GREAT article, and interesting how it applies here too
How I interpreted their ideas was that the scandals regarding bad papers and views was due to a lack of statistcal (math) training and quoted in particular, Biololgists.

Yep...seems like that to me too...
cant learn/reproduce due to the disconnect between the reporting and the data analysis, which are separate formats
It must also be important to train people to follow basic rules and show all work. However, nothing was said about those who read and peer view or how to combat present problems. I don't think ststs training is the answer although it does help

I am with you... may not be THE answer, but it WILL help... and CANT hurt...
just sped thru the article... going back over it later!
(reading Clegg right now)
THANKS for the link! really interesting article!