Project

suika

0.01
No release in over a year
Suika is a Japanese morphological analyzer written in pure Ruby.
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
 Dependencies

Runtime

>= 0.2.0
 Project Readme

Suika

Build Status Gem Version BSD 3-Clause License Documentation

Suika ๐Ÿ‰ is a Japanese morphological analyzer written in pure Ruby.

Installation

Add this line to your application's Gemfile:

gem 'suika'

And then execute:

$ bundle install

Or install it yourself as:

$ gem install suika

Usage

require 'suika'

tagger = Suika::Tagger.new
tagger.parse('ใ™ใ‚‚ใ‚‚ใ‚‚ใ‚‚ใ‚‚ใ‚‚ใ‚‚ใ‚‚ใฎใ†ใก').each { |token| puts token }

# ใ™ใ‚‚ใ‚‚  ๅ่ฉž,ไธ€่ˆฌ,*,*,*,*,ใ™ใ‚‚ใ‚‚,ใ‚นใƒขใƒข,ใ‚นใƒขใƒข
# ใ‚‚      ๅŠฉ่ฉž,ไฟ‚ๅŠฉ่ฉž,*,*,*,*,ใ‚‚,ใƒข,ใƒข
# ใ‚‚ใ‚‚    ๅ่ฉž,ไธ€่ˆฌ,*,*,*,*,ใ‚‚ใ‚‚,ใƒขใƒข,ใƒขใƒข
# ใ‚‚      ๅŠฉ่ฉž,ไฟ‚ๅŠฉ่ฉž,*,*,*,*,ใ‚‚,ใƒข,ใƒข
# ใ‚‚ใ‚‚    ๅ่ฉž,ไธ€่ˆฌ,*,*,*,*,ใ‚‚ใ‚‚,ใƒขใƒข,ใƒขใƒข
# ใฎ      ๅŠฉ่ฉž,้€ฃไฝ“ๅŒ–,*,*,*,*,ใฎ,ใƒŽ,ใƒŽ
# ใ†ใก    ๅ่ฉž,้ž่‡ช็ซ‹,ๅ‰ฏ่ฉžๅฏ่ƒฝ,*,*,*,ใ†ใก,ใ‚ฆใƒ,ใ‚ฆใƒ

Since the Tagger class loads the binary dictionary at initialization, it is recommended to reuse the instance.

tagger = Suika::Tagger.new

sentences.each do |sentence|
  result = tagger.parse(sentence)

  # ...
end

Test

Suika was able to parse all sentences in the Livedoor news corpus without any error.

require 'suika'

tagger = Suika::Tagger.new

Dir.glob('ldcc-20140209/text/*/*.txt').each do |filename|
  File.foreach(filename) do |sentence|
    sentence.strip!
    puts tagger.parse(sentence) unless sentence.empty?
  end
end

suika_test

Contributing

Bug reports and pull requests are welcome on GitHub at https://github.com/yoshoku/suika. This project is intended to be a safe, welcoming space for collaboration, and contributors are expected to adhere to the Contributor Covenant code of conduct.

License

The gem is available as open source under the terms of the BSD-3-Clause License. In addition, the gem includes binary data generated from mecab-ipadic. The details of the license can be found in LICENSE.txt and NOTICE.txt.

Respect

  • Taku Kudo is the author of MeCab that is the most famous morphological analyzer in Japan. MeCab is one of the great software in natural language processing. Suika is created with reference to the book on morphological analysis written by Dr. Kudo.
  • Tomoko Uchida is the author of Janome that is a Japanese morphological analysis engine written in pure Python. Suika is heavily influenced by Janome's idea to include the built-in dictionary and language model. Janome, a morphological analyzer written in scripting language, gives me the courage to develop Suika.