NAME
Treex::Tool::Parser::JDEPP - perl wrapper for C implemented japanese dependency parser JDEPP
VERSION
version 0.13095
SYNOPSIS
use Treex::Tool::Parser::JDEPP;
my $model_dir; # default 'data/models/parser/jdepp/kyoto-partial'
my $parser = Treex::Tool::Parser::JDEPP->new( model_dir => $model_dir );
my @tokens = qw(わたし は 日本語 を 話し ます);
my @tags = qw(名詞-代名詞-一般-* 助詞-係助詞-*-* 名詞-一般-*-* 助詞-格助詞-一般
-* 動詞-自立-*-* 助動詞-*-*-*); # IPADIC tagset
my $parents_rf = $parser->parse_sentence( \@tokens, \@tags );
DESCRIPTION
This is a Perl wrapper for JDEPP Parser implemented in C. This parser works with tokens and POS tags generated by MeCab tagger for dependency parsing. Tokens are grouped together into "bunstetsu", then parsing is performed on these bunsetsu. Simple dependencies between each tokens are generated later in this module.
INSTALLATION
Before installing JDEPP, make sure you have properly installed the Treex-Core package (see Treex Installation), since it is prerequisite for this module anyway. After installing Treex-Core you can install JDEPP using this Makefile (username "public" passwd "public"). Prior to runing the makefile, you must set the enviromental variable "$TMT_ROOT" to the location of your .treex directory.
You can also install MeCab manually but then you must link the installation directory to the ${TMT_ROOT}/share/installed_tools/parser/jdepp/ (location within Treex share), otherwise the modules will not be able to use the program.
METHODS
-
Returns reference to the list of parent nodes for input tokens.
KNOWN ISSUES
So far, JDEPP program could not be succesfully compiled on i686 architecture. If someone will be able to solve this, please send me an e-mail.
SEE ALSO
JDEPP Home Page more info on JDEPP parser
AUTHOR
Dušan Variš <dvaris@seznam.cz>
COPYRIGHT AND LICENSE
Copyright © 2014 by Institute of Formal and Applied Linguistics, Charles University in Prague
This module is free software; you can redistribute it and/or modify it under the same terms as Perl itself.