kanjistat and kanjidist -- new project(s) using KanjiVG

[Dominic Schuhmacher]

First of all, thanks a lot to Ulrich Apel, Ben Bullock and everybody else who is helping to make this project thrive! The KanjiVG data has been my constant companion ever since I've realized that I want to create something that combines my professional expertise (mathematics/statistics) with my private infatuation (Japanese language).

As a first output of this, I would like to propose the following to items as additions to your list of projects that use KanjiVG:

kanjistat - A toolbox for analyzing kanji in the programming language R.
https://dschuhmacher.github.io/kanjistat/
(note also the data package at https://github.com/dschuhmacher/kanjistat.data, which among other things contains R objects derived from (substantially processed) kanjiVG data)

kanjidist - An interactive visualization of kanji distances produced with the above package.
https://www.kanjidist.org

The first one targets a special audience (programmers, statisticians). The second one, I hope, will be useful to kanji learners in general for sorting out confusions between similar looking kanji. Both projects are joint work with Lennart Finke and will be further improved in the future.

Kind regards,

Dominic

[Ben Bullock]

Thank you for letting us know about your projects. I've added both of them to the list at kanjivg.tagaini.net/projects.html. I seem to remember seeing something like kanjidist a while ago but we don't seem to have any links to your website, and you say it's a new project, so it must have been someone else.

 

[Dominic Schuhmacher]

I am not aware of anything similar, so if you remember where you saw this, I'd be definitely interested to know.

The R package started more than a year ago and there are some pictures in https://arxiv.org/pdf/2304.02493 (p.18), which you might have seen somewhere.

Also there is this: https://thekanjimap.com/%E9%9B%A2 (already on your list), which has a somewhat similar vibe, but is very different in that it gives maps of inidividual kanji, not of the relation between kanji.

[Ben Bullock]

On Monday 27 May 2024 at 21:30:25 UTC+9 dominic.s...@gmail.com wrote:

I am not aware of anything similar, so if you remember where you saw this, I'd be definitely interested to know.

It was very similar to yours, it had a kanji with spokes coming out of it to related kanji and you could click on the related elements. I think it was posted on the KanjiVG mailing list, otherwise I honestly don't remember what it was called at all. I've had a bit of a look for it but not done an exhaustive search. I wasn't editing the website at that time, it was Alexandre doing it. It was more than ten years ago as I remember.

 

The R package started more than a year ago and there are some pictures in https://arxiv.org/pdf/2304.02493 (p.18), which you might have seen somewhere.

Also there is this: https://thekanjimap.com/%E9%9B%A2 (already on your list), which has a somewhat similar vibe, but is very different in that it gives maps of inidividual kanji, not of the relation between kanji.

It might have been that, I don't remember it rotating around like that though. I suppose the more things there are, the better though.