I am new to the research domain and recently submitted a survey paper to ACM Computing Survey Journal (CSUR). After the first review, one reviewer commented,

Taxonomy is missing, and ACM CSUR paper must include a taxonomy.

But I have no idea what they mention as an ACM CSUR taxonomy. Looking forward to some support on this.

  • 1
    What is CSUR? And did you mean "one reviewer" instead of "one author"? May 26, 2023 at 15:01
  • 3
    I am almost certain they are asking you to classify your paper: dl.acm.org/journal/csur/author-guidelines#ccs . In nay case, you should ask the editor. May 26, 2023 at 15:28
  • @WolfgangBangerth sorry for the inconvenience. ACM CSUR means ACM Computing Survey Journal. May 26, 2023 at 15:41
  • @AnderBiguri Thank you for the comment. But isn't this after accepting the paper? Currently, it is in the review stage. However, I think it is better to ask the editor as you suggested. May 26, 2023 at 15:47
  • @RoshinieJayasundara yes, but the reviewer can say whatever they want in their review, so I suspect they are referring to that. May 26, 2023 at 15:53

3 Answers 3


This remark about taxonomy is ambiguous. My gut feeling is that the reviewer feels that a CSUR should review the field, which is more than a mere listing of previous work but includes an ordering, i.e. a taxonomy. This is different from classifying the paper itself.

It's like a taxonomy in zoology: There are many animals, but they naturally fall into groups such as mammals, and within mammals canines and felines and then into closely related species and then into species. For example, if you make a survey articles on recent developments of databases, you would need to distinguish between relational and non-relational databases. The latter, sometimes called NoSql databases would fall into (at least) key-value stores, columnar databases, and graph databases. The former would fall into main memory and disk/storage based databases. Your taxonomy should be much better than what I just wrote. For example, object oriented databases and object relational databases are lacking.

The taxonomy should allow the reader to easily place contributions that you did not talk about with contributions that you did talk about. It should allow the reader to easily follow the flow of development. In my example, I could have written a database paper on how to make an in-memory cloud-based database durable. Your taxonomy should allow a reader to immediately identify the key ideas and see how they would work for a graph database.

If my understanding of taxonomy is correct, then it explains the reviewers comment that a surveys article needs a taxonomy. This remark then becomes a rather devastating criticism that needs to be addressed (or refuted).


one reviewer commented, Taxonomy is missing, and ACM CSUR paper must include a taxonomy.

What is the ACM taxonomy?
These two papers, (1) and (2), equate ACM CCS to ACM taxonomy: They are used interchangeably.

Going by the ACM CSUR authors guide (see extract below), the ACM Computing Classification System (CCS) is mandatory.

Content Indicators
Three types of content indicators must be assigned: (1) general terms, (2) subject descriptors, and (3) keywords and phrases. The first two items are selected from the 2012 ACM Computing Classification Scheme. Select as many of these as may be applicable.

I know from experience dealing with ACM publications that it is expected and compulsory.

Purpose: ACM CCS/taxonomy

  • classify manuscript
  • provide proper indexing and retrieval information from the CCS
  • provides readers with quick content reference, facilitating the search for related literature
  • searches in ACM's Digital Library and on other online resources.

Without insight to your manuscript or section of the reviewers comment, it's difficult to say authoritatively in your instance.
The reviewer assumed it's common knowledge for people in computing field used to ACM.

For certainty, you might consider approaching the editor for guidance, just as in this question.

[Edit1] clarify CCS and taxonomy

ACM CCS/Taxonomy are used to classify papers published in ACM journals like the CSUR. They are also engaged, where need be, within a paper (as part of research work): this isn't compulsory.

  • A paper that applies the ACM taxonomy for self-classification does not "include a taxonomy" - it applies/uses one. Jun 26, 2023 at 8:46

Please have a look at the papers in the journal and also IEEE Communications Tutorials and Surveys. You will see that all papers have many mind maps (diagrams, tables, etc) to categorize research topics, problems, tools, etc.

All surveys MUST have such mind maps because they are the most valuable parts of any survey. Otherwise, you have an annotated bibliography that almost anyone can write.

  • Would the wordings all, must, any not be misleading regarding the ACM CSUR(Computing Survey)? Jun 27, 2023 at 6:34
  • 1
    'All surveys MUST have such mind-maps' might be generalising beyond bound. Jun 27, 2023 at 6:36
  • I'm speaking of my area. The papers in that journal require all papers to have a taxonomy. Please read papers from that journal yourself. Hence 'all, must, any' is definitely true.
    – VitaminE
    Jun 27, 2023 at 19:37

You must log in to answer this question.

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