Has anyone written a script to help disambiguate vocabulary with identical English “meanings”?
I’m thinking of something that compiles all the English primary and alternate meanings for every vocabulary item, then looks for items that accept an identical response for “meaning”.
It wouldn’t be hard to write something, but I wondered if it already exists.
I see posts like this one periodically. I’ve written some myself.
It would be interesting to at least view a sorted list of words with identical “meanings” (in quotes because often these words mean different things entirely).
I’m ultimately interested in identifying “hash collisions” where one English phrase has multiple, quite different, connotations, but Japanese has different words for the different connotations.
The first step is to just generate the list. The next (much harder) step would be to identify “hash collisions” vs. actual synonyms.
(Scripts like the Confusion Gueser do more than what I’m looking for.)
So is this the point I recommend starting a “one disambiguation a day” thread?
I don’t think there’s a script like that, at least haven’t seen anyone use anything like it and it would be out of date by now I’m pretty sure
It is probably not what you are looking for, but your description reminded me of the Kanji Search Notes script which lists groups of related words (similar kanji, pronunciations, meanings) for kanji and vocab items.
This would be a (hopefully) fun little forum stat script. I want to know how many of my posts has been liked by any given user.
For example, the number of my posts that have been liked by TofuguJenny or Vanilla are not so many, so those are no problem to count manually. On the flip side, we have Naphthalene who has liked so many of my posts I can’t really count all of them, but at the same time, is not showing up in the “most liked by” column on my profile page (yet).
If this kind of script is possible at all, preferably just a button (which shows the total number of posts) somewhere around the top of the page, like this:
@Redglare I tried making a simple script that searches through the user_actions.json endpoint. This should serve you well, I think. For some reason it seems to be off by a couple of likes compared to the summary page, but it’s pretty accurate