Take the 2-minute tour ×
English Language Learners Stack Exchange is a question and answer site for speakers of other languages learning English. It's 100% free, no registration required.

In English we sometimes use the expression "As its/their name indicates" to express the fact that the name of a concept/object/etc. bears part of its meaning, e.g.:

As their name indicates, supernovae are discovered in the sky as “new stars” (-novae) of exceptionally high brightness (super-).

How would you say in a similar fashion that the name is a misnomer?

For example:

We are going to use logistic regression, which as its name does not indicate is a classification model.

I would like to replace "which as its name does not indicate" by some expression that underlines the fact that the name is a misnomer, not just uninformative.

share|improve this question
    
Not knowing what logistic regression is, I'm not sure if I get your question correctly. Are you looking for despite that (as many below have answered) because the term suggests otherwise than stated? Or, perhaps, are you looking for something implying that the term used is neither telling us what it means nor telling us the opposite? –  Konrad Viltersten Jul 23 at 8:48
    
@KonradViltersten First option: the term suggests otherwise than stated. "logistic regression" is a misleading term, because regression is not classification. Yet, for historical reason, "logistic regression" is still being used. –  Franck Dernoncourt Jul 26 at 15:53

7 Answers 7

up vote 18 down vote accepted

I would use despite:

We are going to use logistic regression, which, despite its name, is a classification model.

share|improve this answer
2  
I was about to add another suggestion "though its name suggests otherwise", but I saw this answer first. I believe this ("despite its name") is much more idiomatic. –  Damkerng T. Jul 22 at 10:22
1  
Alternatively, "in spite of" if it seems that the either the name or the behavior are intentionally opposite. –  Doc Jul 22 at 13:18
    
We always forget about the most obvious, simple choice! –  Franck Dernoncourt Aug 24 at 17:35

I would put it thus:

We are going to use logistic regression, which is, quite contrary to what its name suggests, a classification model.

Or we could use the word counterintuitive:

We are going to use logistic regression, which is a counterintuitive name for what is in fact a classification model.

In place of counterintuitive, one could also use the adjectives misleading and unlikely.

Or:

We are going to use logistic regression, which is a classification model, as its name fails to indicate.

We are going to use logistic regression, which is a classification model actually, although its name fails to put that across.

In place of fails, one could place neglects.

share|improve this answer
1  
I like "misleading" as in: "We are going to use logistic regression. The name is misleading, since it's in fact a classification model. –  PA6OTA Jul 22 at 16:24

I like CopperKettle's suggestions. I think I would write it like this myself:

We are going to use logistic regression, which unlike its name suggests, is actually a classification model.

share|improve this answer
4  
This sounds a bit odd to my ears. Are you a native speaker, and could this be a regionalism from your area? –  David Wallace Jul 22 at 13:00
    
@David, my native language is Australian English, so it shouldn't be substantially different to yours, my Kiwi friend. I couldn't find "(un)like its name suggests" in the COCA or BNC corpora. "Like its name suggests" however is very common in Google Books, and "unlike its name suggests" is not uncommon. –  Dangph Jul 22 at 14:06
    
Yeah, if I read it in an Australian accent, it makes slightly more sense :-) It's certainly not something I'd ever say though. –  David Wallace Jul 22 at 14:08
    
I'm British and this does sound a bit odd. Not "wrong" in the grammatical sense, but not something I think I'd be likely to say. –  GeorgeMillo Jul 23 at 13:26

It seems to me that the problem with most of these suggestions is that the term "logistic regression" does not have an evident surface meaning at all, and in that sense is not obviously problematic. I would say something like: "We will look at a statistical technique which predicts whether an individual has some property or not and which is known, a bit obscurely, as logistic regression"

share|improve this answer
    
"Regression" does have a meaning, or rather meanings (there's regressing to childhood and regressing to a mean), and none of those meanings have anything to do with classification. But +1 for "a bit obscurely". –  Martha Jul 22 at 18:38

We are going to use the { ill-named | poorly named | ineptly named ... } logistic regression, which is actually a classification model.

Here, bold mark the words that work together to express the speaker's opinion about the name. Italics indicate the spoken emphasis: in a contrastive sentence, the elements being compared are stressed. Here, from the emphasis we know that "regression" is being held against "classification" (and not, say, "model"), which is important.

In place of "ill-named" or "poorly named", we can use "so-called", which is less boldly critical of the naming, yet still calls the naming into question, or at least calls attention to it.

share|improve this answer

There is a word that means exactly this:

heterological
(grammar) Of an adjective, not describing itself.

It's constructed as the logical counterpart/antonym of autological, "a word expressing a property which it also possesses itself."

However, this is a pretty rare word, and not all dictionaries list it. So in actual practice, you're better off using some derivative of "contrary".

We are going to use logistic regression, which, contrary to what its name indicates, is a classification model.

We are going to use the self-contradictorily named "logistic regression", which is a classification model.

share|improve this answer

I would omit needless words and let the facts speak for themselves:

We are going to use logistic regression, a classification model.

or,

We are going to use logistic regression as our classification model.

share|improve this answer
    
Obviously, OP feels that making a remark on the naming isn't needless. If people are new to the material, it may help them to draw attention to any naming quirks in the domain so that they avoid confusion. If something called regression really isn't regression at all, it is worth pointing out. You would hardly want to introduce "linear programming", which is a computer science topic, without making it clear that it has nothing to do with software development. –  Kaz Jul 22 at 19:24
    
@Kaz The difference is pointed out by ", a classification model." –  user8993 Jul 22 at 19:25
    
Only to a listener who is "tuned in" to naming issues. –  Kaz Jul 22 at 19:26

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

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