Wikipedia:Articles for deletion/Evolutionary Algorithm for Landmark Detection

The following discussion is an archived debate of the proposed deletion of the article below. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the article's talk page or in a deletion review). No further edits should be made to this page.

The result was Rename. Good work everybody; per the arguments made here I am withdrawing my nomination and moving the page to Landmark detection. Hopefully some of these sources make it in! (non-admin closure) jp×g 11:37, 21 December 2022 (UTC)[reply]

Evolutionary Algorithm for Landmark Detection

[Hide this box] New to Articles for deletion (AfD)? Read these primers!

(Find sources: Google (books · news · scholar · free images · WP refs) · FENS · JSTOR · TWL)

Evolutionary algorithms are real, but there's no indication here that this is a notable application, and the article is of very low quality. A WP:BEFORE search yields squat. Basically no results on Google Scholar -- only one, and it's a list likely scraped from Wikipedia. jp×g 02:13, 7 December 2022 (UTC)[reply]

Note: This discussion has been included in the list of Computing-related deletion discussions. jp×g 02:13, 7 December 2022 (UTC)[reply]
Draft. The subject is notable, but the article is a mess which is why the article should be drafted so other editors can fix the issues wrong with it.`~HelpingWorld~`^(👽🛸) 23:27, 12 December 2022 (UTC)[reply]

Relisted to generate a more thorough discussion and clearer consensus.
Please add new comments below this notice. Thanks, ✗plicit 02:25, 14 December 2022 (UTC)[reply]

~~Keep~~ per WP:NEXIST such as "A Novel Genetic Algorithm for 3D Facial Landmark Localization", "Landmark-based music recognition system optimisation using genetic algorithms" and "Research on genetic algorithm based on tabu search for landmark image recognition". I agree the article is in a poor state but I don't see any purpose in draftifying unless someone is volunteering to work on it. The creator has not been here for twelve years. Sending it to draft would just be slow-motion deletion. SpinningSpark 14:38, 14 December 2022 (UTC)[reply]

@SpinningSpark: I respect the effort here; I hadn't been able to dredge these up. Still, I am not convinced that a standalone article is warranted -- three papers with seventeen citations between them seems like it would make for one or two paragraphs at best. If I am wrong, and these end up being the bulwark of a beautiful article on evolutionary algorithms for landmark detection, I will gladly withdraw my nomination. jp×g 08:33, 16 December 2022 (UTC)[reply]

I didn't pick the example papers for the number of cites. I picked them because they explicitly had both "landmark detection" and "evolutionary|genetic algorithm" in the titles which made them unarguably on topic. But if number of cites is your concern then Automatic Tuning of a Fuzzy Visual System Using Evolutionary Algorithms has 41 cites and included in the text "Landmark detection is a fundamental task in autonomous...". The first paper I linked has 28 cites alone according to gscholar so I don't know how you got to 17 total. SpinningSpark 14:16, 16 December 2022 (UTC)[reply]

@Spinningspark: The sources linked are primary research papers and under Wikipedia:Notability general notability guidelines notability is established through coverage of secondary sources. Even if a secondary source existed, I would argue that this topic is best handled within a section of Evolutionary algorithm and as I'd argue there is nothing of value currently in that article it is best just deleted. EvilxFish (talk) 15:10, 19 December 2022 (UTC)[reply]

You are shifting the goalposts again. First you want sources, then sources with lots of cites. Now, so you say, primary research is ruled out. Nowhere in WP:N does it say that primary research papers don't add to notability. Of course they do. Lot's of people researching a topic is almost the definition of notability in a science subject. We must be cautious how we use primary sources, but they are not proscribed from being used at all, and even if they were, that does not stop them adding to notability. Notability does not fail because the tabloid press has not run sensationalist stories about it. Overview papers are not primary sources and are pretty much de riguer in medical articles. The first sentence of "Evolutionary algorithms for fuzzy control system design" is "This paper provides an overview on evolutionary learning methods for the automated design and optimization of fuzzy logic controllers." It discusses landmark recognition. And it has 188 cites. So what additional requirement are you going to add to rule that one out as well? SpinningSpark 16:58, 19 December 2022 (UTC)[reply]

Not sure how I can be shifting goalposts after my first comment. As for the part in WP:N it states as follows: "Sources" should be secondary sources, as those provide the most objective evidence of notability. There is no fixed number of sources required since sources vary in quality and depth of coverage, but multiple sources are generally expected. Sources do not have to be available online or written in English. Multiple publications from the same author or organization are usually regarded as a single source for the purposes of establishing notability."

Also again, even if we do establish evidence of lots of research covered by secondary sources such as review articles or certain textbooks, I still contest that this is best included within the Evolutionary algorithm page or maybe a more general page on landmark detection, as it is just the application of standard algorithms such as partical swarm and genetic algorithms with maybe a few nuanced points. These algorithms have been applied to a wide range of fields and generally a mention in a more general article, in this case one called "landmark detection", noting any unique features these algorithms have for this specific application, alongside other approaches and link to the general evolutionary algorithms page, is more appropriate - I just don't think the topic is wide enough to justify an article when the information is best presented as part of a more general one. I would argue for a rename or merge if there were anything of value in the article. Now I could be wrong about this, some topics such as Neuroevolution deserve their own article because there are a lot of unique features when applying genetic algorithms to training neural nets, aside from the more general article on neural networks or evolutionary algorithm, even if they use standard algorithms like genetic algorithms as their base. Is the same true here, or is it as I suspect, better covered by a more general article? EvilxFish (talk) 00:11, 20 December 2022 (UTC)[reply]

Sorry for the comment about moving goal posts, I thought it was the same user replying both times. You may be right that this is better covered in a general landmark detection article. There are certainly numerous review papers on the topic [1][2][3][4][5][6][7]. But that is surely an argument for merge and expand rather than delete. Or at worst, rename and cut out the crap if no one can be bothered to work on it. As I've said at AFD before, it is totally perverse to our mission to delete the only information we have on a notable subject because it only covers part of it. We're here to build the encyclopaedia, not stamp on the bits that aren't yet perfect. SpinningSpark 13:32, 20 December 2022 (UTC)[reply]

ok given an article for landmark detection doesn't yet exist, why don't we vote for this one to be renamed to "landmark detection", that way it still exists as a placeholder but is a more general article which will include the evolutionary algorithm approaches as part of it. If you agree with this I will also change my vote to "rename" EvilxFish (talk) 22:54, 20 December 2022 (UTC)[reply]

delete The topic is best covered by a section in the evolutionary algorithms page, also there is a lack of secondary sources covering this topic, suggesting it fails WP:NOTE (but even if there was I would argue it is best covered in the aforementioned article). There is nothing of value in the article in its current state so a merge is not necessary hence I vote delete. EvilxFish (talk) 15:10, 19 December 2022 (UTC)[reply]
rename After discussion with @Spinningspark: I argue this article should be renamed to a more general "landmark detection" which will include the evolutionary approaches as well as other approaches as part of it. EvilxFish (talk) 23:30, 20 December 2022 (UTC)[reply]
Rename to Landmark detection and repurpose. SpinningSpark 00:21, 21 December 2022 (UTC)[reply]
Rename: Per above, I think that this would address my notability concerns (obviously landmark detection is a thing which people do), so when I am back at the computer I will withdraw. jp×g 11:35, 21 December 2022 (UTC)[reply]

The above discussion is preserved as an archive of the debate. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the article's talk page or in a deletion review). No further edits should be made to this page.