18
$\begingroup$

Well, it took a while, but I'm finally done with this effort: every question in has been sorted into and/or .

  • is now empty; a moderator should destroy it.

  • Synonymizing to would remove the slight misclassification of lumping kernel regression in the KDE tag. I didn't count as I read them, but I would very roughly estimate that maybe 15% of questions currently in are actually about kernel regression, and another reasonable portion are about kernel functions in general (not specific to KDE or regression). There are also a handful of questions in , which should also be synonymized to if that's what's happening.

  • The relevant tag wikis need to be updated. Since I don't have direct editing privileges on them, I figured I should wait to propose anything until exactly what's happening with the various kernel smoothing tags is finalized.


Edited to bump. It seems like no one now disagrees with the proposal of making both and into synonyms of , and leaving as is. If anyone does have comments about that, please say so.

$\endgroup$
10
  • 5
    $\begingroup$ Good job. There are only five questions in [kernel-regression], why don't you manually change them to [kde] + [regression] as well? That's probably easier than wait for a moderator action. Also, I am not sure I understand what you suggest about [kernel-smoothing]. Currently this tag does not exist anymore. Regarding the wikis, I think you should go ahead and make the suggested edits! You are definitely the one most familiar with these tags by now, so you are in the best position to write the wiki excerpts. They can always be edited further. $\endgroup$
    – amoeba
    Jul 3, 2015 at 8:32
  • $\begingroup$ I agree w/ @amoeba's points. Go ahead & edit the wiki's; the edits are very likely to be approved. $\endgroup$ Jul 3, 2015 at 15:01
  • 4
    $\begingroup$ @amoeba It's not at all clear to me that kde + regression is adequate -- indeed I think that's actively misleading since kernel regression is NOT kde. kernel-smoothing + regression would be much better. $\endgroup$
    – Glen_b
    Jul 3, 2015 at 17:21
  • $\begingroup$ @gung can you clarify for me how kernel regression is adequately described by kde, as amoeba seems to suggest? Presently that makes no sense to me. $\endgroup$
    – Glen_b
    Jul 3, 2015 at 17:22
  • 4
    $\begingroup$ I agree with @Glen_b on renaming to kernel-smoothing, since there's no density estimation going on in kernel regression, though I may not have said that clearly above.... $\endgroup$
    – Danica
    Jul 3, 2015 at 17:24
  • 1
    $\begingroup$ @Glen_b, you're right about that, sorry. My main point was to go ahead & edit the wikis. $\endgroup$ Jul 3, 2015 at 17:38
  • 2
    $\begingroup$ Yes, my thought was to rename all of [kde] to [kernel-smoothing], as well as the current 5 threads of [kernel-regression]; I renamed into that because I wasn't clear what the outcome of the previous conversation was other than that [kernel-regression] and [kde] don't need to be separate. Re wikis, the only reason I didn't do that yet was because it slightly depends on whether we go with [kde] or [kernel-smoothing]. $\endgroup$
    – Danica
    Jul 3, 2015 at 17:44
  • 2
    $\begingroup$ This discussion seems to have halted. So the Dougal's suggestion seems to be to be to mass-rename [kernel-density-estimate] into [kernel-smoothing], i.e. (?) to make them synonyms with [kernel-smoothing] being the "main" tag, correct? @Glen_b, you are the only moderator taking part in this discussion so far; do you agree with this approach? Can you do it? $\endgroup$
    – amoeba
    Jul 7, 2015 at 14:47
  • $\begingroup$ @amoeba We don't even have an upvoted answer; I'm not sure there's a clear consensus yet. $\endgroup$
    – Glen_b
    Jul 8, 2015 at 3:01
  • 2
    $\begingroup$ @Glen_b: I understand. However, according to my experience of tag-related discussions on meta.SE, it might be easier for Dougal to manually retag all ~200 questions than to reach what will look like a clear consensus here. I recall asking a question on meta.SE with some tag merge suggestions that got 20+ upvotes and still nothing was done and remains so. My impression is that tag changes are very difficult to organize via meta. $\endgroup$
    – amoeba
    Jul 8, 2015 at 9:03

3 Answers 3

4
$\begingroup$

So I'm going to map & to a new .

$\endgroup$
5
  • 2
    $\begingroup$ Great! I'll propose some wiki edits once you do. $\endgroup$
    – Danica
    Oct 15, 2015 at 17:53
  • 1
    $\begingroup$ @Dougal: Thanks - now done. $\endgroup$ Oct 16, 2015 at 12:34
  • 1
    $\begingroup$ @Dougal, there is now several questions (actually only 7) that have [kernel-smoothing] tag and "kernel regression" phrase, but are not tagged with [regression]: see search results. Some of them might have had [kernel-regression] tag before (but after the merge there is no way to know). I guess it would make sense to add [regression] tag to them. Perhaps you can take a look? (I know nothing about this topic myself.) $\endgroup$
    – amoeba
    Oct 16, 2015 at 17:01
  • $\begingroup$ Tag wiki edits proposed, and I'm looking through @amoeba's suggested results to add [regression]. Probably many of the questions now tagged [kernel-smoothing] should also have [density-estimation] (currently barely-used), but I think it's probably fine that they don't. $\endgroup$
    – Danica
    Oct 16, 2015 at 19:40
  • $\begingroup$ @Dougal, great, I took part in approving your tag wiki edits. Very impressive that you actually wrote full wikis (not only excerpts), good job. By the way, consider accepting Scortchi's answer. $\endgroup$
    – amoeba
    Oct 16, 2015 at 20:00
4
$\begingroup$

FYI, I noticed a few weeks ago that had come back from the dead (a little early for Halloween), again. I just removed it from the last question of this batch, but bumping this here as a reminder to active users that the tag is bad and should be killed again if it pops up again.

$\endgroup$
1
$\begingroup$

With regard to blacklisting , please read: How should we make tag blacklist requests?—site mods can't blacklist tags themselves. Noting that hasn't recrudesced, & that SE might well take that as grounds for not blacklisting it, I'd rather not bother. It has, & @amoeba's had to deal with it.

  • Upvote this answer if you agree it's not worth the candle.

  • Downvote this answer if you disagree, & think the dead & buried still needs a stake through its heart.

$\endgroup$
2
  • 4
    $\begingroup$ Given that the procedure is relatively complicated, I am not sure if [kernel] has to be blacklisted or not (so I have not voted either way so far). But I should say that [kernel] did recrudesce: I have recently (perhaps two weeks ago) noticed it and manually edited around 7 questions to replace it. I think it only takes a user with enough rep to recreate it by accident, and it will be used every now and then after that. Perhaps it is enough if the community monitors the situation and terminates it whenever it shows its head again. $\endgroup$
    – amoeba
    Oct 15, 2015 at 14:16
  • 3
    $\begingroup$ I don't have a strong opinion that it needs to be blacklisted, as long as we keep an eye out for it coming back again. If it keeps coming back then we can reconsider. $\endgroup$
    – Danica
    Oct 15, 2015 at 17:14

You must log in to answer this question.

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