JanZerebecki added a comment.

In https://phabricator.wikimedia.org/T117875#1835587, @Cheetah90 wrote:

> > If the AI ingested everything, this would be solved, right?
>
> Yes to some extent, since the AI system still have to figure out which are 
> the correct sub-articles given a main article from all the corpus.


Why would it need to? If the information that chocolate contains caffeine is 
contained in the article "Information that needs to be moved to the correct 
article" but neither in the Chocolate nor Caffeine article then the AI could 
extract that information without knowing anything about sub-article 
relationships.

> False positive: main-article  Caffeine 
> <https://en.wikipedia.org/wiki/Caffeine> sub-article History of Chocolate 
> <https://en.wikipedia.org/wiki/History_of_chocolate>

That is not the sub-article relationship the article implies as is, as it omits 
the sub-article relationship encoded in the headers / table of content of that 
Article. It says Caffeine -> History of Caffeine -> Discovery and spread of use 
of Caffeine -> History of chocolate, which seems to fit your definition, thus 
not a false positive.

> A quick question: how often are the wikidata imported back to Wikipedia 
> articles?

When a Wikipedia article uses 
https://www.mediawiki.org/wiki/Extension:Wikibase_Client#Data_transclusion to 
display information, after that is edited on Wikidata it will under normal 
circumstances update after less than a minute.


TASK DETAIL
  https://phabricator.wikimedia.org/T117875

EMAIL PREFERENCES
  https://phabricator.wikimedia.org/settings/panel/emailpreferences/

To: Lydia_Pintscher, JanZerebecki
Cc: Cheetah90, JanZerebecki, Lydia_Pintscher, Aklapper, StudiesWorld, aude, 
Halfak, Wikidata-bugs, Mbch331



_______________________________________________
Wikidata-bugs mailing list
[email protected]
https://lists.wikimedia.org/mailman/listinfo/wikidata-bugs

Reply via email to