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 no consensus. Dennis Brown - 2¢ 00:31, 7 September 2016 (UTC)[reply]

Fish School Search[edit]

Fish School Search (edit | talk | history | protect | delete | links | watch | logs | views) – (View log · Stats)

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

Yet more nature-inspired metaheuristic cruft. Created August 10th by a SPA with an apparent COI. Conveniently forgets to mention that this algorithm has been criticized for being the particle swarm optimization algorithm under a different name. —Ruud 12:20, 12 August 2016 (UTC)[reply]

(The source I had criticizing this to be equivalent to PSO actually refered to another "fish swarm" metaheuristic. My objections to this article still stand. Statements like "success is represented as weight of each fish" used by Amorim Neto below are exactly the kind of language that e.g. the Journal of Heuristics denounces with "implementations should be explained by employing standard optimization terminology, where a solution is called a 'solution' and not something else related to some obscure metaphor." Independent, high-quality sources are still absent.) —Ruud 16:43, 23 August 2016 (UTC)[reply]

Note to closing admin: Arim Neto and Joaomonteirof, who voted below, are also the authors of the papers this article is based on (and may thus have a slight COI here). —Ruud 12:15, 27 August 2016 (UTC)[reply]

• Comment That's not true, I am under-graduate student and did not write any paper on FSS. I suggest you start reading the papers and based yourself on facts before exposing opinions. There is no COI. Perhaps the misunderstanding is because of "Neto", which is not a family name, it means grandson. Hugo Amorim Neto (talk) —Preceding undated comment added 00:16, 2 September 2016 (UTC)[reply]

Note: This debate has been included in the list of Mathematics-related deletion discussions. Shawn in Montreal (talk) 16:24, 19 August 2016 (UTC)[reply]

• Keep

Fish School Search is an original population based metaheuristic search algorithm first proposed in ^[1].

This family of algorithms, comprised of dozens of versions and variations, uses a totally different search mechanism when compared to PSO. For example, success is represented as weight of each fish (there is no concept of velocity) and change in the total weight of the School. Notice that 'school' is the collective of fish as some people do not know that other meaning for this word. FSS does not use any kind of positional memory (l-,g-best, for example). Very important to notice that FSS does not require any topological knowledge at all, which is absolutely necessary for PSO to be used (i.e. there is connectivity pattern required in FSS). FSS can automatically switch between exploration and exploitation modes, which is something not existing in PSO. Incidentally, this particular FSS operator (catering for automatic search strategy) was included and tested into PSO and produced a much more effective algorithm ^[2]. To conclude, I would like to point out that FSS has now multimodal, multi-objective, binary, parallel etc versions which grant to the FSS family the righteous status of an established role in the big family of Swarm Intelligence SI techniques.

We suggest that further and careful read be pursued in the references so as the above pointed out differences (there are more interesting ones) clearly set FSS apart, as an original brand new family of algorithms with SI. Therefore, it is rather unfair to cast FSS as a copy of anything in literature.Hugo Amorim Neto (talk) 19:04, 20 August 2016 (UTC) — Hugo Amorim Neto (talk • contribs) has made few or no other edits outside this topic. [reply]

• Keep

FSS utilizes totally different search mechanisms when compared to PSO! — Preceding unsigned comment added by Joaomonteirof (talk • contribs) 12:54, 23 August 2016 (UTC) — Joaomonteirof (talk • contribs) has made few or no other edits outside this topic. [reply]

• Keep I think we should keep these metaheuristics articles until the scientific argument is truly settled. It is quite easy to wait several years before the community decides that that these optimization algo's are not subsets of the PSO. I know it's a niche community, particularly related to limited sources pool, but it is good enough at the moment for WP. If it doesn't work out, they can be very easily removed, in say, 5 years, when it is fully settled. Scope creep (talk) 18:41, 25 August 2016 (UTC)[reply]

Note: This debate has been included in the list of Computing-related deletion discussions. Lemongirl942 (talk) 12:32, 27 August 2016 (UTC)[reply]

• Comment This is another of the Swarm intelligence algorithms, isn't it? The original paper seems to have 65 citations (which is quite low considering the field) and I don't know how many are self citations. Unlike the Ant and Bee Colony optimisation algos, this one was proposed in 2008 but doesn't seem to have caught on. I am leaning towards a delete as I am unable to find that the idea is a groundbreaking one which deserves a Wikipedia article at this time. --Lemongirl942 (talk) 12:41, 27 August 2016 (UTC)[reply]

Note: This debate has been included in the list of Science-related deletion discussions. Lemongirl942 (talk) 12:41, 27 August 2016 (UTC)[reply]

That sums up the situation pretty well, yes. —Ruud 21:21, 27 August 2016 (UTC)[reply]

• Delete as WP:Too soon with insufficient sources. Too few citations in GS as yet. Xxanthippe (talk) 22:25, 27 August 2016 (UTC).[reply]

• Comment 1. The metaphor free approach for Metaheuristcs is still a matter of opinion of some researchers. There is not any consensus in Academia defining whether we should or not use metaphors to describe algorithms; 2. I've got here a list with 59 FSS based papers published in international Journals/Conferences by researchers from many countries. Is it relevant?; 3. What are the formal relevance criteria for considering an article to be published in Wikipedia? — Preceding unsigned comment added by Joaomonteirof (talk • contribs) 18:32, 6 September 2016 (UTC)[reply]

• Keep If this article is deleted we have also to delete all the other articles which don't have a "relevant" number of citations in GS or all the metaphor based metaheuristic algorithms such as PSO, GA, ABC, among many others. --Isabela Maria Albuquerque (talk) 19:51, 6 September 2016 (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.