4

I am currently writing my Ph.D. thesis. The thesis is in computer science specifically the data science domain. The thesis will be based on my 8 first-authored publications in IEEE, ACM and springer. The final thesis will be eventually available online for public by the university library.

Can I use the text and figures I created and published in the conferences papers? to what extent can I copy from them? will it be considered self-plagiarism if I do so?

Looking into some of my previous colleagues' Ph.D. published thesis, I found that they have used the same exact text and figures without even citing their original IEEE papers(they mentioned in the introduction a list of papers references that are used in the thesis) while some created complete new figures to represent the results and some just included the figures while citing their papers in the captain.

PS1: I Wrote the copyrights@ieee.org and will update the question if I get a reply.

PS2: The answer from IEEE is (That is indeed acceptable, as IEEE allows authors to use everything short of their entire paper for thesis reuse without requiring permission, provided that everything is cited. If you need to use an entire paper however, you will need to contact our permissions department (pubs-permissions@ieee.org). The following URLs may be of use to your for this and future post-publication work, and the first link specifically deals with thesis reuse.

https://journals.ieeeauthorcenter.ieee.org/choose-a-publishing-agreement/avoid-infringement-upon-ieee-copyright/

https://journals.ieeeauthorcenter.ieee.org/become-an-ieee-journal-author/publishing-ethics/guidelines-and-policies/post-publication-policies/

https://conferences.ieeeauthorcenter.ieee.org/author-ethics/guidelines-and-policies/post-publication-policies/ )

2
  • 2
    You will need to consult your local institutional policy to determine whether reusing text from prior publications is permitted, or whether the thesis needs to be all new at time of submission. That varies, and without that information people can only be speculating. Feb 14 at 19:06
  • It would be good if you posted your edit (and @MichaelHomer, perhaps your comment) as an answer instead.
    – wizzwizz4
    Feb 14 at 19:08

4 Answers 4

1

You need to consult your institutional regulations with regard to the self-plagiarism question. Nobody here can help you with that.

Requirements for doctoral theses vary, and span the full range of

  • The thesis work must not have been published previously in any way.
  • You can reuse all the text, but have to edit it into a coherent whole.
  • You literally staple three papers together and hand it in.

and many points in between. You may not be allowed to do this at all, or it may be the standard way a thesis is written where you are.

While everybody thinks that what they're used to is universal, only someone who knows your actual institution's rules can answer that question usefully for you. If you've been seeing it happen plenty, presumably it's fine - but you want to know where the line is.


On the copyright issue, both the IEEE and the ACM permit anywhere up to total reuse of the contents of accepted papers within a thesis or dissertation, including figures, as long as a citation to the version of record of the paper is included. This is an explicit part of the copyright agreement that you make when publishing the work for virtually all legitimate publishers in one form or another (sometimes by reference to a published policy).

For the ACM:

Authors can include partial or complete papers of their own (and no fee is expected) in a dissertation as long as citations and DOI pointers to the Versions of Record in the ACM Digital Library are included.

For the IEEE:

You may reuse your published article in your thesis or dissertation without requesting permission, provided that you fulfill the following requirements ... [fine-grained citation format rules for text and figures follow]

(You can infer from the existence of this blanket licence term that such theses are common, but they're not universal)

If "they mentioned in the introduction a list of papers references that are used in the thesis" and inside captions, your colleagues have likely satisfied that requirement. The IEEE also requests copyright symbol markers virtually everywhere, though to-the-letter observance of that is limited in my experience.


Generating new figures from previous data sets may well be outside of the copyright transferred to the publisher and not require their permission at all, but it will depend on the nature of the figures. It's not required in order to reuse them within your thesis, but you might prefer to have them a consistent style throughout anyway. You can cite them with "adapted from ..." if it's derivative.

6

There are actually two issues, copyright and self-plagiarism.

The latter is easy to avoid if you quote and cite the earlier work in the thesis. Cite it as you would cite the work of another. This holds for both the text and the images.

The copyright issue can be a bit more involved, but likely is not. When you give copyright to a publisher you normally get back a license for certain uses. Among the typical permitted uses is for a dissertation. You may have the specific wording of your license for reuse somewhere in you email (or paper mail), but both IEEE and ACM will have them online if you look. This license for reuse is typical of all reputable publishers, not just those that are also professional societies. But, you should quote and cite even here. But longer quotations than are typically allowed (work of others) is almost certainly ok but not unlimited.

Your colleagues should have done this also.

3

Although many things related to intellectual properties are a grey area, the only certain thing is that copyrights do not cover the pixel of a figure, or the wording in a sentence, copyrights cover the intellectual property.

I found that they have used the same exact text and figures without even citing their original IEEE papers while some created complete new figures to represent the results

While the first group that used same text and figures without citing IEEE papers probably broke the copyright (maybe they submitted the thesis before IEEE papers?), the second group definitely broke the copyright, because they knew they were representing the same intellectual content!

Stay away from shortcuts, ask the permission for reproducing the figures/plots/etc. or simply cite your own work with something like "see Figure 7 from Krebto (2020d)" without including the figure.

The final thesis will be eventually available online for public by the university library.

Yes, but I am 99.9999% sure that when you upload the final PDf to the library for publication, you will sign some document stating that you have all the rights and you obtained permission to reproduce any contents included in the thesis. Since you are asking the question, it seems you do not want to declare something you know is false.

1
  • 3
    Both the publishers mentioned provide a blanket licence to reuse paper contents in your thesis, and all of the uses identified in the question clearly comply with that. This doesn’t seem to answer the question at all and in parts seems outright incorrect. Feb 14 at 19:03
1

I think you can include figures or text from IEEE papers with citation. To be on the safe side, ask the editors of the journal, if you need a permission.

For more explanation, please read here.

2

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .