Advertisement
MO BIO
MO BIO

Plagiarism detection 2.0

Publishers are getting a new tool in the fight against plagiarism in scientific manuscripts. The Scientific business of Thomson Reuters linkurl:announced;http://scientific.thomsonreuters.com/press/2008/8452130/ on May 1 that they would be offering their clients - the publishers of many well-read science journals - the option to employ iThenticate, a tool that checks submitted manuscripts for potential copy-catting against databases of previously published work. According to Logan Hutchinson,

By | May 12, 2008

Publishers are getting a new tool in the fight against plagiarism in scientific manuscripts. The Scientific business of Thomson Reuters linkurl:announced;http://scientific.thomsonreuters.com/press/2008/8452130/ on May 1 that they would be offering their clients - the publishers of many well-read science journals - the option to employ iThenticate, a tool that checks submitted manuscripts for potential copy-catting against databases of previously published work. According to Logan Hutchinson, a product manager in the Scientific business at Thomson Reuters, iThenticate "will reduce linkurl:research integrity violations;http://www.the-scientist.com/article/home/38023/ and linkurl:plagiarism";http://www.the-scientist.com/news/20040922/02/ by using technology to point editorial staff members in the direction of such infractions. "It's meant to provide a body of evidence," Hutchinson said. The iThenticate tool will be offered as a part of the Manuscript Central service that many of Thomson Reuters' publisher-clients use for automated manuscript submission. iThenticate will digitally search, or "crawl," the text of submitted manuscripts, comparing it to existing manuscripts stored in publisher-stocked databases, such as linkurl:CrossCheck,;http://www.the-scientist.com/blog/display/54546/ or to public libraries of papers, such as linkurl:PubMed.;http://www.the-scientist.com/news/20050512/02/ Searching the public-facing internet as well as CrossCheck will help the program will flag more "fingerprinted" passages that show similarity to published papers that editors can check to see if plagiarism has occurred. "iThenticate is the technology and software that is powering all of the crawling and fingerprinting," Hutchinson said. Hutchinson said that iThenticate will be fully integrated into Manuscript Central "sometime in 2009," with the CrossCheck database set to come online this June. He also said that iThenticate will help editors to catch plagiarism early on in the submission and peer review process, rather than after a plagiarized manuscript has been published. Journals that employ Manuscript Central, and will have the option of using iThenticate, include the __New England Journal of Medicine__, the linkurl:American Chemical Society;http://www.the-scientist.com/news/home/53719/ and linkurl:Oxford University Press;http://www.the-scientist.com/article/home/53781/ families of publications, and journals published by Blackwell Publishing, such as __Acta Zoologica__, __Clinical Genetics__ and __Marine Ecology__. __(Editor's Note: The original version of this story listed the journal__ Cell __as one of publications using Manuscript Central. This is not the case, and mention of the journal has been removed from the story.__ The Scientist __regrets the error.)__
Advertisement
Keystone Symposia
Keystone Symposia

Comments

Avatar of: ROBERT HURST

ROBERT HURST

Posts: 31

May 13, 2008

I don't know how prevalent the practice is, but I know of two papers published by a multi-investigator consortium that have almost identical abstracts and similar titles. I suspect the motivation was to increase the first and last authorships of members of the consortium. While I do understand the motivation, the papers were so similar as to bring into question whether one paper plagiarized the other. In addition I know it is common practice for research groups to slice and dice their results and publish several papers that are very similar to each other just to increase their apparent productivity, or so it seems. This kind of checking should reduce these practices considerably. The net result is to clog the literature with duplication.
Avatar of: anonymous poster

anonymous poster

Posts: 10

May 13, 2008

What a ridiculous idea! Are we dealing with college students? Who will pay the extra time for the editors and reviewers needed? Are we going to add another assurance at the end of every article: "The authors assure that we have not committed any plagiarism and never used more than 10 words (insert #) from any other article, without reference given". \n\nAnd what are we going to do with the sinners? And who is going to be the judge? And where is the court system that will handle the transgressions so that normal legal rules of defending, appeal, independent judge etc are fulfilled? In short where is common sense?
Avatar of: Ellen Hunt

Ellen Hunt

Posts: 199

May 13, 2008

There is a special plagiarism that bright students, (and maybe others) particularly in bioengineering and such, commit with impunity. They search through patent office filings, (generally prior to issue of patent) to find something to use for a thesis. Then they duplicate it using the directions in the filing. \n\nSearching for this needs to be made part of plagiarism automated systems.
Avatar of: anonymous poster

anonymous poster

Posts: 1

May 13, 2008

I would have thought that quoting yourself without attribution was not plagiarism even though it may be undesirable duplication. After all, can you steal from yourself? But it appears from other posts that this opinion is not universal.\n\nSo, can I cut and paste from myself without being accused of theft and at what point is duplication a crime?
Avatar of: Pascale Lane

Pascale Lane

Posts: 3

May 13, 2008

In general this is a good idea, but the idea of plagiarizing from myself is a bit absurd. After all, I publish multiple studies, all using the same model. How many ways can I describe that model? In every paper, there are 1 or 2 paragraphs that are virtually identical because the process is identical. If this tool flags items such as this, it will be more of a pain than an aid!
Avatar of: mukesh varshney

mukesh varshney

Posts: 1

May 14, 2008

It is really a positive effort to stop plagiarism. \nwhat are the limitations?\nsometimes we do same techniques for different hypotheses. In that case the description of the technique is same which the author has published earlier. How many ways one has to edit the description for the same protocol or technique. Would it be plagiarism if one describes the text from earlier of his publications for the same technique or method.\nThis problem needs to be defined very clearly.
Avatar of: Inna Sokolova

Inna Sokolova

Posts: 1

May 14, 2008

While I completely agree that publishing the same data twice is bad (and irritates me whenever I come across such semi-identical articles), it does not sound from the software description that it will efficiently prevent this. I suspect that it will be mostly finding repetitive descriptions of the experimental procedures and such in Materials and Methods sections that move unchanged from one article to another (and rightly so - why invent a new description of the same method if the authors, reviewers and editors had agreed that the previous one is clear and well written) and occasional paragraphs from the intro that explain why the study topic is important and the model is appropriate (there are only so many ways to say the same thing). Unless the software has a capability of filtering out such occurrences, the signal-to-noise ratio for its plagiarism detection will be very low and it will place a huge (unnecessary) burden on the editors to manually screen all the alleged plagiarisms, 90% of which will likely turn out to be the normal practice of describing the same procedure in the same words within the limitations of the technical language. Even Tolstoy and Hemingway tended to use the same phrases, words and constructions in their writing which is the basis of how linguists can ascribe authorship to questionable texts; is that also plagiarism? \nAnd by the way, I think if the authors who have no ethical scruples about publishing their data twice want to do it, they still will be able to - all they will need is to be more creative in their writing and move the words around.
Avatar of: anonymous poster

anonymous poster

Posts: 26

May 14, 2008

All dictionary definitions of Plagiarism I found specify that it is dupicating *some one else's* words or art without permission. Thus, by definition, one cannot self-plagiarize.\n\nSelling the same item as an "orginal" more than once is a different kind of fraud....\n\nBaxter Zappa

Follow The Scientist

icon-facebook icon-linkedin icon-twitter icon-vimeo icon-youtube
Advertisement
Panasonic
Panasonic

Stay Connected with The Scientist

  • icon-facebook The Scientist Magazine
  • icon-facebook The Scientist Careers
  • icon-facebook Neuroscience Research Techniques
  • icon-facebook Genetic Research Techniques
  • icon-facebook Cell Culture Techniques
  • icon-facebook Microbiology and Immunology
  • icon-facebook Cancer Research and Technology
  • icon-facebook Stem Cell and Regenerative Science
Advertisement
New England BioLabs
New England BioLabs
Advertisement
NeuroScientistNews
NeuroScientistNews