Japanese NLP - Introductions

Languages at Hugging Face
#1

This is the introduction thread for Japanese NLP practitioners.

Welcome! Please introduce yourself and let us know:

  • Your name, Github, Hugging Face, and/or Twitter handle
  • Your interest in Japanese NLP
  • Some projects you are working on or interested in starting
  • Any other languages that you speak, any personal interests, anything else really :wink:
4 Likes
#2

Iā€™m Yusuke Mori.
My Github account: forest1988

My interest in Japanese NLP:
Although I mainly work with English texts as my research subject, my first language that I use daily is Japanese.
I was involved in the Japanese translation project of spacy-course. I would be pleased if I could make some contribution in the ā€œLanguages at Hugging Faceā€.

Projects I am working on or interested in starting:
Iā€™m now working on the documentation of BERT Japanese.

Any other languages:
I am not very good, but I can communicate in English to some extent.

5 Likes
#3

Hello, my name is Paul Oā€™Leary McCann. Iā€™m the creator of fugashi, the MeCab wrapper which is used for Japanese tokenization in some BERT models. I also maintain mecab-python3, which was used in Transformers before fugashi. I also have done a lot of work on the Japanese models in spaCy and worked on improving speed in SudachiPy last year.

My interest in Japanese NLP:

I live in Japan and have worked in Japanese industry for years, and as an independent consultant the past year and a half. I think that while there are good tools for working with the unique challenges Japanese presents thereā€™s often issues with ease of use or maintenance, so a lot of my open source work has focused on that.

Projects I am working on or interested in starting:

I just released Kanji Club, a kanji search site, though it isnā€™t a machine learning project. Other than that, Iā€™m not working on anything in Transformers actively at the moment but I enjoy keeping up with community developments and look forward to seeing what people come up with! Please do feel free to @ me if you have any trouble with or questions about the MeCab wrappers.

Actually, if the Livedoor News Corpus isnā€™t in datasets yet it should probably be added, so thatā€™s a project ideaā€¦

Some of my projects:

  • Kanji Club: Just released instant kanji search-by-parts site
  • fugashi: Pythonic Cython-based MeCab wrapper
  • cutlet: A library to Romanize Japanese

Elsewhere online:

5 Likes
#4

Hello, Iā€™m Atsunori Fujita.

My interest in Japanese NLP:
I would like to add more Japanese language models so that Japanese developers can easily build Japanese language solutions.

Some projects you are working on or interested in starting:
I am planning to add a Japanese pre-training model(Considering XLNet, Roberta).

Building datasets and developing tokenizers are crucial concerns in my planning. I would be happy to cooperate with you

2 Likes
#5

The documentation of BERT Japanese is included in the v4.6.0 release!

1 Like
#6

Hello, I am Kazuma Takaoka.
I am a developer of Japanese morphological analyzers, Sudachi and SudachiPy.
We have recently started a project to use SudachiPy as a tokenizer for transformers, and we are planning to release models using Sudachi.

We welcome your comments on our project.

1 Like