[PYTHON] [Morphologische Analyse] So fügen Sie Mecab ein neues Wörterbuch hinzu

#Hatena-Schlüsselwort
curl -L http://d.hatena.ne.jp/images/keyword/keywordlist_furigana.csv | iconv -f euc-jp -t utf-8 > keywordlist_furigana.csv
# Wikipedia
curl -L http://dumps.wikimedia.org/jawiki/latest/jawiki-latest-all-titles-in-ns0.gz | gunzip > jawiki-latest-all-titles-in-ns0

require 'csv'

original_data = {
  wikipedia: 'jawiki-latest-all-titles-in-ns0',
  hatena: 'keywordlist_furigana.csv'
}

CSV.open("custom.csv", 'w') do |csv|
  original_data.each do |type, filename|
    next unless File.file? filename
    open(filename).each do |title|
      title.strip!

      next if title =~ %r(^[+-.$()?*/&%!"'_,]+)
      next if title =~ /^[-.0-9]+$/
      next if title =~ /Begriffsklärung/
      next if title =~ /_\(/
      next if title =~ /^PJ:/
      next if title =~ /Figuren/
      next if title =~ /Aufführen/

      title_length = title.length

      if title_length > 3
        score = [-36000.0, -400 * (title_length ** 1.5)].max.to_i
        csv << [title, nil, nil, score, 'Substantiv', 'Allgemeines', '*', '*', '*', '*', title, '*', '*', type]
      end
    end
  end
end

ruby sample.rb

/usr/local/libexec/mecab/mecab-dict-index -d /usr/local/lib/mecab/dic/ipadic -u custom.dic -f utf-8 -t utf-8 custom.csv

$ sudo vi dicrc

userdic ="Speicherort des Verzeichnisses des erstellten Wörterbuchs"

#coding:utf-8
import MeCab
tagger = MeCab.Tagger("-Ochasen")
result = tagger.parse("Wolke")
print result

Kura Kura Kura Nomen-Proprietäre Nomenklatur-Allgemeines
Udo Udo Nomen-Allgemeines

Cloud Cloud Substantiv-Allgemeines

Recommended Posts