10

Background:

In order to form a body of work to show to prospective employers, I am implementing a genetic solver for the weighted maximum independent set problem. The code will be available on github under an open-source license. While researching the problem, I came across this paper. However, I found it to be lacking justification for many of its decisions. Attempting to gain a deeper understanding of the topic, I went through other papers that it cited as references. I quickly found Beasley and Chu's A Genetic Algorithm for the Set Covering Problem which provides a much clearer explanation of many of the same issues.

In fact, upon closer reading, I realized that many paragraphs from the first paper were copied--almost word for word--from the second one (e.g. the last paragraph in section 3.6 from Beasley and Chu is identical to the first paragraph on page 8 of the other paper). Perhaps most damning is the reproduction of the heuristic-feasibility operator with absolutely no attribution to the original authors. After discovering this, I stopped referencing the derivative paper (which, by the way, greatly sped up my development process).

Question:

First, is the first paper guilty of plagiarism? If so, how much of this should I discuss in the code and the project summary on github? Should I even mention the derivative paper? I don't want to slander the author in a public forum--although this case seems pretty straightforward--and I certainly don't want to commit plagiarism, but I feel some obligation to the original authors to point people to their work instead of the derivative one.

  • 3
    Why not just cite the earlier (and if I understood right, more comprehensible) paper in the first place? – dmckee Jun 22 '15 at 1:12
  • Sorry if I was unclear, first refers to the first paper I mentioned, the possibly plagiarized one. I avoided using the authors name or the title for obvious reasons, but the link is there. – Dylan MacKenzie Jun 22 '15 at 1:23
  • 1
    It's entirely appropriate to just cite the earlier paper if that contains all the ideas you are using. – Jim Conant Jun 22 '15 at 1:43
  • If they quote someone three times and cite them twice, I would suspect that the third was not intentional. Either way, I can't understand why you would want to cite a quote when you have access to the original? – WGroleau Dec 8 '17 at 1:45
13

I suggest you to be very careful to label the first paper a plagiarism, unless you are 100% sure that there are areas of a word-for-word equivalence with the original paper(s) without credit for original authors. Having said that, I noticed that the first paper's author cite the second one (authored by Beasley and Chu) - and another one by the same authors - as well as clearly state that the same (or similar) approach has already been used by Beasley and Chu (references #22 and #23), so it seems that some credit has been given (which is not to say that some is enough).

In regard to which paper(s) to reference, I would say that it depends on the following two aspects: 1) if the second paper is indeed a case of plagiarism, then it certainly doesn't make sense to cite it; 2) if it is not, then the decision should be made, based on whether there are distinctly different sets of ideas in both papers and whether you have used both sets or one.

  • 2
    I am just a lowly undergraduate but there are several paragraphs which are appropriated directly from Beasley and Chu without citing them. This is in addition to identical structure and methodology. Although the original author cites Beasley and Chu twice, the instances I am referring to are distinct from the cited passages. That said, I am not a part of the academic wold, and don't feel entirely comfortable passing judgment on someone else's work. – Dylan MacKenzie Jun 22 '15 at 1:35
  • 6
    @DylanMacKenzie: If that is the case, then I would just cite the original paper (by Beasley and Chu). In addition, I think that it is appropriate to send the corresponding author (Beasley) a note, informing about the lack of full credit in the derivative paper - since it appears that that paper is not published, there is a chance that this situation can be fixed by revising the paper accordingly. – Aleksandr Blekh Jun 22 '15 at 1:47
  • That seems like a good solution. I realize that this is not the best forum for these sorts of questions, but I was unsure how else to proceed – Dylan MacKenzie Jun 22 '15 at 1:55
  • 8
    @DylanMacKenzie: I think that Academia.SE is the best forum for such questions. – Aleksandr Blekh Jun 22 '15 at 2:18

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.