For a university project I want to classify authors according to the subjects/domains they frequently publish in. For this purpose I want to first establish a subject classification with the according hierarchies (for example, chemical engineering would be a sub-subject of physical sciences and engineering).

Instead of creating my own classification/hierarchy, I would like to use an official classification of subjects. Unfortunately, I have not been able to find one that is universally accepted.

What I did find are the classifications of different publishers, for example the subjects listed for ScienceDirect.

So my question is:

Is there any official, widely used and accepted subject classification?

One thing I'd like to add: English is not my mother tongue so it might be entirely possible that I am just using the wrong words for doing the search. I’ve been looking for subject or domain classifications. Maybe there is an entirely different term for what I am actually looking for.

  • "Official" according to whom? There is no office that could enforce such a standard. Sep 6, 2017 at 9:29
  • @David Richerby I don't think there needs to be one single office that enforces the standard globally. I meant official in the sense that an organisation/institution published it (and depending on the organisation also uses it). I think the systems put forward by Peter K. fit this discription. In order to exclude classifications that are published but not really used I've added the "widely used and accepted" part. If you think my use of the word "official" is wrong or misleading feel free to suggest a better alternative :) Sep 6, 2017 at 9:57

3 Answers 3


In addition to the ones @Coder suggests, perhaps the library classification systems:

might be useful?

As @tonysdg says below, there are a variety of other system, too.

  • 1
    I hadn't thougt about using library classification systems for this purpose. It seems like a very good approach. Sep 5, 2017 at 16:53
  • 2
    There are a variety of library classification systems available. You may find that a more universal one (such as the aptly-named Universal Decimal Classification system) is of interest.
    – tonysdg
    Sep 5, 2017 at 20:15
  • @tonysdg : Thank-you! I've added your links to the body of the answer.
    – Peter K.
    Sep 5, 2017 at 20:22
  • 3
    These systems have the added benefit for the OP that the authors' works will already have been sorted by cataloging experts into the appropriate categories (though note that authors themselves may have published in multiple different classes and sub-classes).
    – 1006a
    Sep 5, 2017 at 21:05

These are the possible classification that you could look at:

  • Google Scholar --> Metric (a link on the top of the page) --> scroll down to 'view top publications' --> Click categories --> ... explore as per your requirement.

  • ACM subject classification (ACM ccs)

  • Mathematics subject classification by AMS (Wiki)
  • PACS Regular Edition from AIP publishing
  • 2
    In physics, on is the American Institute of Physics (for key words in journal articles): publishing.aip.org/publishing/pacs/pacs-2010-regular-edition
    – Jon Custer
    Sep 5, 2017 at 13:16
  • +1. Thank you very much @JonCuster; I added this link to the answer. I am mostly in CS, so I probably lacked the details.
    – Coder
    Sep 5, 2017 at 18:52
  • 2
    The granularity of the ACM classification can be frustrating. It often feels like single topics are conceded entire hierarchies of classification terms, while on other occasions, several large umbrella topics are mingled together in a single term. To be fair, that is probably a natural result both of historical developments and of people who update the classification being interested only in a subset of all fields covered by it. The OP should, however, consider this if they plan to derive anything such as "number of authors per topic" from their author-to-topic mapping. ... Sep 5, 2017 at 21:53
  • 2
    ... Another, minor issue with the ACM classification is that I wouldn't bet any money that publications are generally consistently classified when looking at the terms that can mean the "meta-level". For instance, I am never quite sure whether "Empirical studies in HCI" should cover all publications that contain such a study, or just such publications that discuss the study methodology, or rather such publications that focus on presenting a new methodology for conducting, or a new technique that supports such studies. Sep 5, 2017 at 22:01
  • I strongly agree with your comments @O.R.Mapper. the acm classification is not entirely acceptable, because of the fact that the articles are sometimes classified by author even if they use a smallest term of the field. It mostly is not acceptable if the article appears in conference proceeding.
    – Coder
    Sep 6, 2017 at 4:50

Although a bit obsolete, consider also the UNESCO nomenclature. It is international and specifically adapted to research. Also, unlike other systems, it is not restricted to Science and Engineering. For instance, in Spain, you have to specify up to three codes to officially register a PhD Thesis.

You must log in to answer this question.

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