Last modified: 2013-10-08 23:09:44 UTC
It could be made a special page that analyzes the clustering of articles and how it reflects on other languages. The most similar of the articles on other languages could then be used to identify the most likely item in Wikidata. This would be a great help in identifying the correct relationship for pages that is otherwise difficult to assign to the correct item.
The special page will use a sitelink as accessor, aka [[Special:LikelyItems/SomeSiteLink]], and list the likely items in descending order with likelihood, label and description.
This kind of page is a maintenance page that aids in creating items.
A better way to use this information would be to manipulate the sort order of the items in the drop down list so the most common yet unused items are on top. If the user writes some characters the list is reduced to those items that has those prefixed characters.
Good idea but I'm closing this to unclutter the bug tracker. Not useful to keep around atm.
