Apertium-tat

From Apertium
Revision as of 08:53, 1 September 2014 by Francis Tyers (talk | contribs)
Jump to navigation Jump to search

Apertium-tat is a morphological analyser/generator and CG tagger for Tatar, currently under development. It is intended to be compatible with transducers for other Turkic languages so that they can be translated between. It's used in the following language pairs:

Current State

{{#set_param_default | corpus1 | None }} {{#set_param_default | corpus2 | None }} {{#set_param_default | corpus3 | None }} {{#set_param_default | corpus4 | None }} {{#set_param_default | corpus5 | None }} {{#set_param_default | corpus6 | None }} {{#set_param_default | corpus7 | None }} {{#set_param_default | corpus8 | None }} {{#set_param_default | corpus9 | None }} {{#set_param_default | corpus10 | None }}

  • Number of stems: 55,702 {{#ifneq | | | () }}
  • Disambiguation rules: 123
  • Coverage: ~91%

{{#ifneq | quran | None |

{{#ifneq | | | | }}

}}

{{#ifneq | NewTestament | None |

{{#ifneq | | | | }}

}}

{{#ifneq | aytmatov | None |

{{#ifneq | | | | }}

}}

{{#ifneq | wp2013 | None |

{{#ifneq | | | | }}

}}

{{#ifneq | tatnews2005/11 | None |

{{#ifneq | | | | }}

}}

{{#ifneq | {{{corpus6}}} | None |

{{#ifneq | | | | }}

}}

{{#ifneq | {{{corpus7}}} | None |

{{#ifneq | | | | }}

}}

{{#ifneq | {{{corpus8}}} | None |

{{#ifneq | | | | }}

}}

{{#ifneq | {{{corpus9}}} | None |

{{#ifneq | | | | }}

}}

{{#ifneq | {{{corpus10}}} | None |

{{#ifneq | | | | }}

}}

corpuswordscoverage
<nowinter>[[|quran]]</nowinter>quran165K ~89.2%
<nowinter>[[|NewTestament]]</nowinter>NewTestament137K ~94.2%
<nowinter>[[|aytmatov]]</nowinter>aytmatov5K ~93.4%
<nowinter>[[|wp2013]]</nowinter>wp2013128K ~87.3%
<nowinter>[[|tatnews2005/11]]</nowinter>tatnews2005/114.6M ~90.7%
<nowinter>[[|{{{corpus6}}}]]</nowinter>{{{corpus6}}} ~%
<nowinter>[[|{{{corpus7}}}]]</nowinter>{{{corpus7}}} ~%
<nowinter>[[|{{{corpus8}}}]]</nowinter>{{{corpus8}}} ~%
<nowinter>[[|{{{corpus9}}}]]</nowinter>{{{corpus9}}} ~%
<nowinter>[[|{{{corpus10}}}]]</nowinter>{{{corpus10}}} ~%

Installation

apertium-tat is located in languages module.

You will need HFST, lttoolbox, apertium and vislcg installed on your computer to be able to use it.

If you are using a Debian-based distro, the easiest way to get those dependencies is to install them with apt-get from User:Tino Didriksen's repository:

wget http://apertium.projectjj.com/apt/install-nightly.sh -O - | sudo bash
sudo apt-get -f install locales build-essential automake subversion pkg-config \
 gawk apertium lttoolbox libapertium3-3.3-dev liblttoolbox3-3.3-dev apertium-lex-tools \
 cg3 hfst libhfst36-dev

Then you can check out apertium-tat from our svn repository and compile it:

svn co http://svn.code.sf.net/p/apertium/svn/languages/apertium-tat/
cd apertium-tat
./autogen.sh
make

Otherwise, see the Installation page for instructions.

Usage

Morphological analysis:

$ echo "барабыз" | apertium -d . tat-morph 
^барабыз/бар<v><iv><pres><p1><pl>$^./.<sent>$

apertium -d . tat-tagger will give disambiguated output, apertium -d . tat-disam -- disambiguated output in CG format showing which rules were applied:

$ echo "Без урманга таба барабыз." | apertium -d . tat-disam 
"<Без>"
	"без" prn pers p1 pl nom
;	"без" prn pers p1 pl nom
;		"и" cop p3 pl REMOVE:236
;	"без" prn pers p1 pl nom
;		"и" cop p3 sg REMOVE:236
"<урманга>"
	"урман" n dat
"<таба>"
	"таба" post SELECT:543
;	"таба" n nom
;		"и" cop p3 pl REMOVE:236
;	"таба" n nom
;		"и" cop p3 sg REMOVE:236
;	"таба" n attr REMOVE:452
;	"тап" v tv prc_impf SELECT:543
;	"таба" n nom SELECT:543
;	"тап" v tv pres p3 sg SELECT:543
"<барабыз>"
	"бар" v iv pres p1 pl
"<..>"
	".." sent

If you want to know what exactly each mode does (in our case, modes are "tat-morph", "tat-tagger" and "tat-disam"), have a look at the modes.xml file or the modes/ directory.

Components

  • apertium-tat.tat.lexc : lexicon (stems + morphotactics)
  • apertium-tat.tat.twol : morphophnological rules
  • apertium-tat.tat.rlx : disambiguation rules