Prototype of Classical Chinese Morphological Analyzer based on MeCab

Morphological analyzer is very important to process large scale text
database.  It is not only used for linguistic analysis, but also it is
used as a basis of full text search and various text processing.  For
major languages, morphological analyzer and other tools about
linguistic analysis are available, however for minor languages,
especially for classical languages, few tools are available.
classical Chinese is an example.  Morphological analyzer requires
dictionaries and tagged corpus, and it is very hard to develop them
from scratch.  If prototype of classical Chinese Morphological
Analyzer is available, it may be helpful to develop tagged corpus and
dictionaries.  So I tried to make a prototype of classical Chinese
morphological analyzer based on MeCab, which is a language-independent
morphological analysis engine.  To use MeCab for classical Chinese, we
need definitions of parts of speech, dictionary and tagged corpus for
MeCab.  To reduce costs, I tried to convert modern Japanese dictionary
into pseudo dictionary for classical Chinese as a prototype
dictionary.  Modern Japanese and classical Chinese share large
vocabulary (in this sense, distance between modern Japanese and
classical Chinese seems nearer than distance between modern Chinese
and classical Chinese), we can use some heuristics to detect classical
Chinese morphemes from modern Japanese morphemes.  Output of MeCab can
be used as tagged corpus for MeCab, and dictionary can be generated
from tagged corpus, so the prototype can be used to develop corpus and
real dictionary.