ZPar statistical parser. Universal language support (depending on the availability of training data), with language-specific features for Chinese and English. Currently support word segmentation, POS tagging, dependency and phrase-structure parsing.
C++ Python Other
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
Makefile.d
doc
scripts
src
.cproject
.gitattributes
.project
.pydevproject
.travis.yml
Makefile
Makefile.ccg cross platform fix: use gmkdir on Windows, MinGW Oct 10, 2014
Makefile.common
Makefile.doc2snt
Makefile.misc
Makefile.zpar.en
Makefile.zpar.es
Makefile.zpar.ge BUGFIX: remane zpar->zpar.zh, zpar.ge->zpar Oct 10, 2014
Makefile.zpar.mvt
Makefile.zpar.zh
README

README

COMPILE
=======
make zpar.en for the English zpar for PTB data
make zpar.zh for the Chinese zpar for CTB data
make zpar for the generic, language and- and tagset-independent system 

make postagger for the generic postagger
make depparser for the generic dependency parser
make conparser for the generic constituent parser

make segmentor, chinese.postagger, chinese.depparser and chinese.conparser for the word segmentation system, postagger, depparser and conparser for Chinese, respectively.

make english.postagger, english.depparser, english.conparser, english.ccgparser for the postagger, dependency parser, constituent parser and CCG parser for English, respectively. 

Both language specific versions of ZPar have hard coded tagsets.