Is there a data structure that can sensibly handle multiple hierarchical classification systems?
e.g. an Orange, in terms of phylogeny is
Plantae->Eudicot->...->Citrus->sinensis
and in terms of usefulness, is
Thing->Food->fruit->orange
(and it could have multiple parents in this taxonomy, e.g. cleaning product)
Bonus points for cool visualisations of this kind information.
[#]data #dataScience #dataStructure #information #hierarchy #taxonomy #classification #visualisation #dataViz
=> More informations about this toot | More toots from naught101@mastodon.social
@naught101
pretty sure this is exactly what SQL is for, you'd just have a separate table for
Things -- listing all types of things,
Food -- listing the types of food, etc.
There's different ways you could organize the database using unique id's or human readable names, but basically everything is a table and they're all associated with each other, and then the information could be displayed any way you want in html or curses or whatever... i'm coming from a php minset so it would just be a bunch of php echo commands based on the variables.
=> More informations about this toot | More toots from big_louse@todon.eu
@big_louse True, anything is possible in a relational database. It could reasonably be done in nosql or a deductive database. Probably a deductive database or a graph databse make the most sense.
I guess I was wondering if there was already something more specifically designed for this type of problem, but I guess that probably doesn't make sense.
=> More informations about this toot | More toots from naught101@mastodon.social
@naught101
yeah there's a million ways to do, it's basically a set of arrays, what i would do is organize the data in sql and then print the data using xml or html to show that hierarchy, like really basic xml:
orange apple
then style it with css so all the fruits are clustered together etc.
=> More informations about this toot | More toots from big_louse@todon.eu
@big_louse yeah.. I think I probably need to start from the visualisation design FIRST, and then try to create specs for the data requirements from that. Especially because there are some other aspects that I want to include that are potentially orthogonal to a multiple hierarchy..
=> More informations about this toot | More toots from naught101@mastodon.social
@naught101 not a bad idea, that will help you anticipate issues that aren't immediately apparent,, that's how i do my little php web apps, i make a sample html page of what i want the end result to look like, and then i work backwards from that markup writing all the functions etc.
on the db thing tho, doesn't have to be that complicated, sqlite is probably all you need
happy coding
=> More informations about this toot | More toots from big_louse@todon.eu
@naught101 It seems a job for a graph database like neo4J
=> More informations about this toot | More toots from yawmoght@mastodon.social
@yawmoght Damn, that looks cool! Thanks!
=> More informations about this toot | More toots from naught101@mastodon.social
@naught101@mastodon.social Isn't this just a hierarchical tag system? You'd have an object, the orange. It would have multiple sets of tags, like #thing/food/fruit and #plantae/eudicot/.../citrus. A sufficiently advanced search tool will let you look up any tier of tag and get anything tagged with a child of said tag. I'm particularly thinking of the obsidian tag system here.
=> More informations about this toot | More toots from rfc6214@mk.absturztau.be
@rfc6214
Yes, but I'd like more specificity. I guess I was thinking though, is how to define what kind of taxonomy each parent relationship sits in. So how is "citrus" parent different from "fruit" parent? I think things like RDF can handle this, by specifying the exact type of the relationship (not just a generic "parent"), which is just one way of describing parallel taxonomies.
=> More informations about this toot | More toots from naught101@mastodon.social
@rfc6214
And funny you should mention obsidian.. I use logseq a lot, and made a related feature request recently..
https://discuss.logseq.com/t/semantically-meaningful-links-e-g-providesevidencefor-climate-change/6684/4
=> More informations about this toot | More toots from naught101@mastodon.social
@naught101 This was one of the core features of the TOFLIT18 project at @medialab_scpo : http://toflit18.medialab.sciences-po.fr/
Basically this was made by @paulanomalie & @Yomguithereal using @neo4j and classification tables.
=> More informations about this toot | More toots from boogheta@piaille.fr
@boogheta @naught101 @medialab_scpo @Yomguithereal @neo4j Indeed concurrent hierarchical classifications are music to my ears. To learn more you can watch this talk https://archive.fosdem.org/2019/schedule/event/graph_french_trade_study/ and/or read this paper https://www.tandfonline.com/doi/full/10.1080/01615440.2022.2032522
Oh and feel free to ping me if you want to chat about that.
=> More informations about this toot | More toots from paulanomalie@piaille.fr
@paulanomalie @boogheta @medialab_scpo @Yomguithereal @neo4j
Thanks, will have a listen!
=> More informations about this toot | More toots from naught101@mastodon.social
@naught101 Not sure what youโre looking. First thing coming to mind would be relational databases? One table would be phylogeny, another commonUses, etc. Orange in both of them for a join?
=> More informations about this toot | More toots from JauneBaguette@framapiaf.org This content has been proxied by September (3851b).Proxy Information
text/gemini
