Support Japanese morphological analyzer

[nobu-k]

We need to have it.

I don't know what is the right format of the result, though. It can simply be ['word', 'word', ...]. However, each word could have additional information such as its position in the sentence, length, and something like a part of speech 品詞. So, it may be [{'word': 'word1', 'begin': 10, 'end': 15, 'pos': 'noun'}, {...}, ...].