Jump to content

Wikifunctions:Type proposals/Japanese verb conjugation class

From Wikifunctions

Summary

The "classes" or "categories" which denote the patterns of verb conjugation in Japanese (Z1830).

Uses

For NLG obviously.
A verb Lexeme can have multiple Forms attached for its conjugated forms, but if those are incomplete or if there's a need for a verb that's missing from Wikidata entirely, knowing the lemma form and conjugation class would let you derive the other forms.

The impetus for this proposal was that I was thinking about reimplementing this set of enwikt templates.
Current usage is that the caller has to know and pass in the lemma, kana (phonetic) spelling, and conjugation class. With Lexemes, the caller would pass an LID, and those things would be fetched from the lemma representations and statements respectively. I can proceed with Z6091-typed parameters, but an enum would be better on any general-use helper Functions.

Elements of the enumeration

I'm not a native speaker, so this may not be the most idiomatic ordering, and I may have missed an irregular verb which is still used in modern Japanese. The Wikibooks page I referenced has more than these three irregular classes, including some fossil words used for politeness, but it seems in Wikidata those are all non-verbs or regular verbs.

Alternatives

I think this smaller set:

...would also be sufficient, as the subclass can then be inferred from the lemma form. But I went with the above since those are the ones actually in use with conjugation class (P5186) on Lexemes, meaning they could be used directly and it saves that little bit of computation.

Comments

For general comments, please reply to the proposer.

@Higa4 I noticed you were working on Japanese NLG, could you check whether I've made any mistakes here? YoshiRulz (talk) 22:25, 1 April 2026 (UTC)reply
Support Support @YoshiRulz I checked our modeling and it looks fine to me. Thanks a lot! Higa4 (talk) 01:11, 2 April 2026 (UTC)reply