Retraction Watch

Tracking retractions as a window into the scientific process

Fake citations plague some Google Scholar profiles

with 14 comments

lw20

One of the “articles” that was linked to Bucci’s Scholar profile last week (click for full-size)

Last week, microbiologist Enrico Bucci emailed us with concerns that several of the citations listed on his Google Scholar profile were fake.

Colleagues of his had noticed the same problem on their pages.

The listings seem to be real titles, researchers, and publications, but scrambled. When Bucci first spoke with us, the Scholar citations all linked to clearly fake pages on a site hosted by e-commerce giant Alibaba. You can see an example here (that’s a screenshot on the right).

Google hasn’t responded to a request for comment from Retraction Watch, but since we contacted them, the links have been disappearing, replaced by unlinked citation notices (you can see a screenshot of one below).

fake citation

Here’s Bucci’s theory:

Out of the blue, my Google Scholar Profile included some publication which was never authored by me…These citations were actually fabricated by someone removing the true authors and attaching my name (and those of some coauthors of mine) to a true paper (abstract, title and journal are ok; the year is sometime wrong)​.

It appears that Google Scholar is taking this citation as genuine – and pushing it in my Google Scholar profile (I think they use automatic indexing, relying mostly on coauthors for disambiguation of author names).

There is a clear intent of making money out of it. The site hosting these “faked” papers sells the pdf versions of them. There is a page containing the description and the conditions for this service which is here. http://www.lw20.com/Agreement.aspx

Try Google Translate on it.

At this point, I think they are grabbing the pdf papers from publishers, the indexing from google scholar (albeit with some mismatches in author names), then they translate abstract and titles in Chinese – so that Chinese scholars may find the papers – and try to sell the papers via their platform.

Of course, this wouldn’t be the first time someone tried to game Google Scholar.

We’d like to know how deep this problem runs. Do you have any fake references in your Scholar profile? Let us know in the comments.

Update, 3:40 p.m. Eastern, 11/17/14: Several scientists have reached out to tell us about their own experiences with fake citations.

Geologist Anne Jefferson tells us that her Google Scholar profile is accurate, but a number of fake papers with the same titles as her real publications are still available at http://lw20.com/, the Alibaba site that is hosting the odd chimeras. She told us:

I think your correspondent’s theory is correct. Lw20 is scraping the journals, quite poorly and then in some cases google scholar is getting tricked by the mismatched authors into thinking there is more than one paper.

Shark researcher and blogger David Shiffman had one of the fake papers show up on his profile, but he deleted it:

It was super easy. You right click on it and say “remove” and check a confirm box.

 

Written by Cat Ferguson

November 17th, 2014 at 2:22 pm

Posted in faked data

Comments
  • Eibl November 17, 2014 at 3:11 pm

    I have no fake citations on my google scholar site – but if I remember correctly, one should select the correct citations and therefore exclude the wrong ones (usually from other authors with similar names). But this doesn’t solve the mentioned problem of fake (china-related?) citations… will be interesting to see how this problem will be solved.

  • Former Technician November 17, 2014 at 3:40 pm

    I found fake publications on my boss’s google scholar listings two years ago. I added one to his publications list and he said it wasn’t real. The authors listed and the title were very appropriate for the ongoing projects. I just thought it was a publication that I had missed. I then looked at more with similar author combinations and found 5 more. What tipped us off was that one of the authors only published 2 papers in her time with us, but was listed on several both as first and co-author.

    Considering all of the other garbage in google scholar, we have long since stopped using their metrics. Patents, conference abstracts and more showed up. While this is useful to some, it dilutes the h-index by making the publication list look longer.

    My boss is a well established full professor and it hurts him to pad his list with papers that are not able to be cited. I would suspect that it would hurt junior researchers too since academia is looking at publication metrics as part of promotion requirements.

  • Sylvain Bernès November 17, 2014 at 3:45 pm

    With a quick check on my account, I detected one bogus citation:

    Ortiz, A., Quintero, L., Hernández, H., Maldonado, S., Mendoza, G., Bernes, S…. & Shiga, A.
    “DD Wisnoski and CW Lindsley microwave technology”.
    Tetrahedron Letters, 44(6).

    On the other hand, I realize that, for any reason, ca. 50% of my citations were removed from my account.

    • Sylvain Bernès November 17, 2014 at 4:06 pm

      Mhhh. I just spotted a second odd citation… Something strange is happening, or what?
      Should I plan to remove my account?

  • Alberto November 17, 2014 at 4:08 pm

    This sounds like old school spam to me. I don’t think it is much of a problem in Google Scholar Citations, where authors can easily delete any documents (as David Shiffman has already noted), and even select an option to get notified every time the system finds a document that might have been written by them, so they can review it before it is added to their profile.

    However, it could pose a problem if these results are displayed as results in the search engine, where a user might fall for this ruse. In any case, Google Scholar can easily fix this problem by not indexing more papers from this domain (and others like it) and deleting the ones that have already gotten through.

  • maria November 17, 2014 at 4:12 pm

    That lw20.com site has a ton of papers on it – looking for random topics I know of generated a lot of results. It does look like someone is potentially hosting pdfs of papers which may not be easily accessed by Chinese students/scholars and potentially monetizing this ? Not sure. Perhaps a Chinese scholar can take a look. While it’s not affecting my Google Scholar profile (and it wouldn’t matter, I am not an academic), it’s clearly the case that such “citations” could mistakenly be added by Google to one’s profile if the information is different enough from that of the real paper. One can of course manage the profile aggressively, but this looks more like a site which should be blacklisted by Google when retrieving paper data.

  • Peter Schneider November 19, 2014 at 4:05 pm

    Well, if lw20.com hosts unauthorized copies or translations of scientific papers, there is another option. A DMCA complaint (maybe I am wrong). However any publisher should have an interest, that this lw20.com website does not appear in the Google/Google Scholar search results. https://support.google.com/legal/troubleshooter/1114905?hl=en
    I maybe recommend that you contact copyright departments at Elsevier, Wiley, etc. so actions can be taken. Without actions, any other cheap content aggregator such as academia.edu will republish your papers too.

  • Kenneth Witwer November 19, 2014 at 9:58 pm

    One of my papers was listed three times on Google Scholar: once correctly, and a second and third time with the title replaced by or switched with the names of one or more of the co-authors. The result was an inflated citation count, with some articles apparently “citing” more than one version. To remove the duplicates, I “merged” the three, another option on Scholar. I didn’t delve too much into why the problem cropped up, but perhaps it has something to do with how Scholar reads different citation formats.

    • Scholarrific November 25, 2014 at 10:56 am

      I find it more annoying and insidious when my papers are cited in “reputable” journals as support for an argument that is the polar opposite of my own data and/or conclusions. Because many authors are too lazy to read citations before listing them in bibliographies, I’m starting to see those errors/deceptions propagate into other papers. I fear that at some point I will have to start writing to journals or posting on PubPeer to clarify that the citation misrepresented my original work.

      Writing the authors citing my work has led nowhere. Any suggestions as to how to deal with this quickly and effectively would be most helpful.

      • Marco November 25, 2014 at 11:54 am

        One suggestion is to cc the Editor on your communication with the authors. So far that has helped me get at least two corrections of papers citing one of my papers in support of an equation that I actually strongly criticize as being inappropriate.

        And if it gets really bad, you write a scathing commentary and submit it to one of the journals. Good example: Britta Stordal’s “Citations, citations everywhere but did anyone read the paper?”
        http://www.sciencedirect.com/science/article/pii/S0927776509001490

  • JATdS November 19, 2014 at 11:46 pm

    Peter Schneider, I think you have touched upon an extremely important aspect that has the potential to sway citations massively: content aggregators. May I suggest that we discuss the problems associated with LinkedIn, Academia.edu and Research Gate, etc. please, as I feel that there is a strong overlap with Google Scholar. But I am not sure how such aggregators work, so we need some insight about how papers and citations are linked to other papers and citations, and names, and how this information is aggregated. For exmaple, among my e-mail addresses, receive a really irritatingly amoung of invitations to join these aggregators, but of course I don’t because my CV would be extremely distorted if I did.

  • Dr. Nill January 19, 2015 at 5:15 am

    A curious case of a mistaken author name in a Springer journal:

    Applied Biochemistry and Biotechnology October 2014 Date: 08 Oct 2014
    Effect of Metal Ions and Redox Mediators on Decolorization of Synthetic Dyes by Crude Laccase from a Novel White rot Fungus Peniophora sp. (NFCCI-2131)
    Shiv Shankar, Shikha Nill
    DOI: 10.1007/s12010-014-1279-2
    http://link.springer.com/article/10.1007%2Fs12010-014-1279-2

    The erratum states:
    “The original version of this article unfortunately contained a mistake. The last name “Nill” of “Shikha Nill” should be removed. The correct name is also provided here.”
    http://link.springer.com/article/10.1007/s12010-015-1482-9

    • Lee Rudolph January 19, 2015 at 8:05 am

      Evidently Dr. Shikha uses only one name (see her official university resume); an earlier publication in the same journal by the same two authors, “Laccase Production and Enzymatic Modification of Lignin by a Novel Peniophora sp.”, cited in the references of this paper, got it right. My guess is that some Springer software insists every field of a form must be filled in.

  • Shmuel September 15, 2017 at 6:19 am

    Good day,

    As far as I understand, even if the author removes one of the papers, it will still count in the citation indexes.

    Google says that :

    “Your “Cited by” counts come from the Google Scholar index. You can change the articles in your profile, but citations to them are computed and updated automatically as we update Google Scholar.”

    Therefore, the author’s changing the articles would not affect the metrics.

    Am I correct?

    Thank you!

  • Post a comment

    Threaded commenting powered by interconnect/it code.