Reference · Pairs of kanji that look alike
LOOK_ALIKEconstant
10093 pairs of kanji that look alike, each with scores out of 100: by how few stroke edits separate them, and by
the parts they share (0 where the measure did not rank the pair among either kanji's ten nearest). Kept at 60 or more by
strokes, or 70 or more by parts. The 2882 pairs those two left out (981 of their kanji are ones he never scored) have a
fourth number, the score by shape, made from KanjiVG's strokes, at 75 or more (70 for a kanji he never scored); the
others have three. Written by scripts/build-data.mjs from data-sources/look-alike/. Never edited by hand.
The scores by strokes and by parts are Lars Yencken's, from https://lars.yencken.org/datasets/kanji-confusion/, under the Creative Commons Attribution 3.0 Unported licence: Yencken, Lars (2010) Orthographic support for passing the reading hurdle in Japanese. PhD Thesis, University of Melbourne, Melbourne, Australia. The scores by shape are this package's, made from the stroke paths of KanjiVG (https://kanjivg.tagaini.net, Ulrich Apel and its contributors, Creative Commons Attribution-Share Alike 3.0, https://creativecommons.org/licenses/by-sa/3.0/). The table is derived from both and is under CC BY-SA 3.0.
Import it
import { LOOK_ALIKE } from "@johnmorrisdotca/jisho/look-alike";Definition
const LOOK_ALIKE: PackedLookAlikesMembers
| Name | Type | Description |
|---|---|---|
source | string | Whose scores they are. |
pairs | readonly (readonly [string, number, number] | readonly [string, number, number, number])[] | The two kanji written together, then the score by strokes and the score by parts, each out of 100, 0 where not scored, and the score by shape where there is one (the pairs the first two did not hold). |
Examples
import { LOOK_ALIKE } from "@johnmorrisdotca/jisho/look-alike";
console.log(Object.keys(LOOK_ALIKE));[ 'source', 'pairs' ]What it prints (run when this page was built)