| Japanese-English in Temple
|
Japanese Morphology
The Japanese morphological analyzer is to be incorporated in
Temple as a part of the general morphological analysis functionality.
The glossary look-up and lexicon look-up functions work on the text
annotated by the result of the morphological analysis system, Juman, which has been
developed at Kyoto University under supervision of Dr.Nagao.
To illustrate, consider the following sentence:
"私は昨日学校へ行った。" Juman's first pass produces the
analysis:
| 私 | (わたくし) | 私 | 普通名詞 |
| は | (は) | は | 副助詞 |
| 昨日 | (きのう) | 昨日 | 普通名詞 |
| 学校 | (がっこう) | 学校 | 普通名詞 |
| へ | (へ) | へ | 格助詞 |
| 行った | (おこなった) | 行う | 動詞 | 子音動詞ワ行 | タ形 |
| 。 | (。) | 。 | 句点 |
Notice the sixth row,"行った" for "行う" as a stem (or basic form)
instead of "行く" which is the right analysis. Juman will produce all
possible analyses if you choose to do so.
The third column of Juman's output, a stem or basic form, is used for
an entry into our dictionary database in order to obtain its
translations. Also the fourth column is used for the part of speech of
the word in our system.
Computer Analysis of Japanese Morphology
For the purpose of computer analysis, Juman's output were slightly
modified to fit into our system. First of all, the fourth column of
Juman's output was translated to English: 普通名詞 -> noun, 動詞 ->
verb, and so on.