PANTERA

The PANTERA is a Brill Tagger for morphologically rich languages.

PANTERA's homepage is here.

Author: Szymon Acedański <s DOT accek AT SPAMFREE gmail DOT com>

Downloads

For some distributions it is available in packages:

1. Ubuntu, Mint (as a PPA): ppa:bartosz-zaborowski/nlp (ppa's page on launchpad)

   • (they work at least in Ubuntu 11.04 and Mint 11, for other versions try rebuilding from source package)
2. Fedora 16:
   • i686

     • pantera-0.9r150-1.i686.rpm

     • pantera-debuginfo-0.9r150-1.i686.rpm

     • pantera-lib-0.9r150-1.i686.rpm

   • x86_64

     • pantera-0.9r150-1.x86_64.rpm

     • pantera-debuginfo-0.9r150-1.x86_64.rpm

     • pantera-lib-0.9r150-1.x86_64.rpm

   • source rpm (may work for other versions of fedora too)

     • pantera-0.9r150-1.src.rpm

3. Opensuse 12.1
   • i586

     • opensuse_pantera-0.9r150-1.i586.rpm

     • opensuse_pantera-lib-0.9r150-1.i586.rpm

   • x86_64

     • opensuse_pantera-0.9r150-1.x86_64.rpm

     • opensuse_pantera-lib-0.9r150-1.x86_64.rpm

   • source rpm (may work for other versions of Opensuse too)

     • opensuse_pantera-0.9r150-1.src.rpm

The above packages depend on Morfeusz-SGJP. Binaries for Ubuntu/Mint are available from the same PPA as pantera. RPMs for Fedora and Opensuse are here:

• i686

  • morfeusz-SGJP-0.7-1.i686.rpm

  • morfeusz-SGJP-debuginfo-0.7-1.i686.rpm

  • morfeusz-SGJP-lib-0.7-1.i686.rpm

• x86_64

  • morfeusz-SGJP-0.7-1.x86_64.rpm

  • morfeusz-SGJP-debuginfo-0.7-1.x86_64.rpm

  • morfeusz-SGJP-lib-0.7-1.x86_64.rpm

• source rpm

  • morfeusz-SGJP-0.7-1.src.rpm

Any questions/bugreports/feature requests related to the tagger should be sent directly to the author (the email address is on the homepage). Questions/bugreports related to the binary packages may be sent to the maintainer of packages (bartosz.zaborowski-no-spam@ipipan.waw.pl).

Publications

The tagger is described in a following paper:

• S. Acedański, "A Morphosyntactic Brill Tagger for Inflectional Languages," in Advances in Natural Language Processing, 2010, pp. 3-14.

It is a rewrite of an experimental tagger described in

• S. Acedański and K. Gołuchowski, "A Morphosyntactic Rule-Based Brill Tagger for Polish," in Recent Advances in Intelligent Information Systems, Kraków, Poland, 2009, pp. 67-76.

What does PANETRA stand for?

The acronym PANTERA comes from "Polskiej Akademii Nauk Tager Ekstrahujący Reguły Automatycznie", which means in English "Automatic Rule Extraction Based Tagger of the Polish Academy of Sciences". The word "pantera" means "leopard".