Hello,
No, there is nothing from our side regarding extracting data from Wiktionary. This is not in the plans of the development team, by the way, we think that this decision (to extract or not) and the ways to possibly do it, should be taken by both of the communities (Wikidata and Wiktionary).
If you have any experiments or demo, feel free to share :)
On 20 March 2017 at 21:57, Amirouche amirouche@hypermove.net wrote:
Héllo all!
Le 02/03/2017 à 10:34, Léa Lacroix a écrit :
Hello Amirouche,
Thanks a lot for your interest in this project and your proposal to help. Currently, the development team is still working on the new datatype structure for lexemes, and we don't have something to demo yet.
I don't need wikibase support of L, F and S right now.
What I am wondering is whether there is already work done wikimedia side regarding the *extraction* of Lexeme, Form and Sens from wikitionary pages.
I started scrapping english wiktionary. I will have demo ready by the end of the week. But I'd like to avoid duplicate work and focus on other stuff if wikimedia already plan to do this.
As soon as we can provide a viable structure to test, we will announce it
here and on the talk page of the project https://www.wikidata.org/wiki /Wikidata_talk:Wiktionary.
Cheers,
On 1 March 2017 at 22:43, <fn@imm.dtu.dk mailto:fn@imm.dtu.dk> wrote:
Hi, It is my understanding that Wikidata for Wiktionary requires new data structures or at least new name space (L, F and S), and that is what holding people back. What could be interesting to have would be a prototype (not necessarily built with MediaWiki+Wikibase) to see if the suggested scheme is ok On 03/01/2017 10:16 PM, Amirouche wrote: Héllo, I have been lurking around for some month now. I stumbled upon the wiktionary in wikidata project via for instance this pdf https://upload.wikimedia.org/wikipedia/commons/6/60/Wikidata
_for_Wiktionary_announcement.pdf https://upload.wikimedia.org/wikipedia/commons/6/60/Wikidat a_for_Wiktionary_announcement.pdf
Now I'd like to help. For that I want to build a bot to achieve that goal. My understanding is that a proof of concept of the page 11 of the above pdf can be good. But I never really did any site scraping. Is there any abstraction that help in this regard.
Wikidata mailing list Wikidata@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata