2 # This is a sample user dictionary for Kuromoji (JapaneseTokenizer)
4 # Add entries to this file in order to override the statistical model in terms
5 # of segmentation, readings and part-of-speech tags. Notice that entries do
6 # not have weights since they are always used when found. This is by-design
7 # in order to maximize ease-of-use.
9 # Entries are defined using the following CSV format:
10 # <text>,<token 1> ... <token n>,<reading 1> ... <reading n>,<part-of-speech tag>
12 # Notice that a single half-width space separates tokens and readings, and
13 # that the number tokens and readings must match exactly.
15 # Also notice that multiple entries with the same <text> is undefined.
17 # Whitespace only lines are ignored. Comments are not allowed on entry lines.
20 # Custom segmentation for kanji compounds
21 日本経済新聞,日本 経済 新聞,ニホン ケイザイ シンブン,カスタム名詞
22 関西国際空港,関西 国際 空港,カンサイ コクサイ クウコウ,カスタム名詞
24 # Custom segmentation for compound katakana
25 トートバッグ,トート バッグ,トート バッグ,かずカナ名詞
26 ショルダーバッグ,ショルダー バッグ,ショルダー バッグ,かずカナ名詞
28 # Custom reading for former sumo wrestler
29 朝青龍,朝青龍,アサショウリュウ,カスタム人名