Free natural language morphology

Free morphological analyzer Majka

Binaries

Download Majka binaries for Linux / for Windows.

For the Windows OS the requested encoding is Windows 1250.

$ chcp 1250

Usage: program expects one entry (word, lemma, or string lemma:tag, according the data file in use) per line on its standard input and prints the requested information on its standard output. An example of usage (for other options see majka -h):

$ echo test | majka -f majka.w-lt
test:k1gInSc1
test:k1gInSc4
test:k1gMnSc1
testa:k1gFnPc2

Source codes

• Source code of the analyzer for Linux / for Windows. The code is inspired by Jan Daciuk's fsa package.
• Source code of the program which converts lexicographically sorted data to an automaton. The code is allmost only a simplification of Jan Daciuk's fsa_build.
• All source codes are licensed under GPL version 2.

References

When using majka for research purposes, please cite:

Pavel Šmerk. Fast Morphological Analysis of Czech. In Petr Sojka and Aleš Horák. Proceedings of Third Workshop on Recent Advances in Slavonic Natural Language Processing, RASLAN 2009. Brno : Masaryk University, 2009. p. 13–16. ISBN 978-80-210-5048-8.

Free morphological databases for Majka

• Czech Free, version 2.0 (tagset reference)

  • data for assigning lemmata and tags to analyzed word forms,
  • data for generating all word forms and tags of a given lemma,
  • data for generating word forms according to a given lemma and tag.
  • You may also use our web service API to obtain the information.
  • source vendor: Natural Language Processing Centre, Faculty of Informatics, Masaryk University, Brno, Czech Republic
  • LICENSE: Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License

• Czech Limited, version 2.0 (tagset reference)

  • data for assigning lemmata and tags to analyzed word forms,
  • data for generating all word forms and tags of a given lemma,
  • data for generating word forms according to a given lemma and tag.
  • Data cover 90 % of CZES corpus (around 500 M words). They contain 3,393,080 wordform+lemma+tag triplets, which is 903,888 distinct word forms and 46,000 lemmata. The data also contain parts of almost 8 billion compound words.
  • You may also use our web service API to obtain the information.
  • For the full version of data, please, contact us (see below).
  • source vendor: Natural Language Processing Centre, Faculty of Informatics, Masaryk University, Brno, Czech Republic
  • LICENSE: Creative Commons Attribution-ShareAlike 4.0 International License

• Slovak (tagset reference)

  • data for assigning lemmata and tags to analyzed word forms,
  • source vendor: Radovan Garabík, Ľ. Štúr Institute of Linguistics, Slovak Academy of Sciences
  • LICENSE: Creative Commons Attribution-ShareAlike 3.0 Unported License , GNU AFFERO GENERAL PUBLIC LICENSE, version 3 and GNU Free Documentation License

• Polish (tagset reference)

  • data for assigning lemmata and tags to analyzed word forms,
  • source vendor: PoliMorf
  • LICENSE: 2-clause BSD license

• Swedish (tagset reference)

  • data for assigning lemmata and tags to analyzed word forms,
  • source vendor: Språkbanken
  • LICENSE: Creative Commons Attribution-ShareAlike 3.0 Unported License and GNU Lesser General Public License

• German (tagset reference)

  • data for assigning lemmata and tags to analyzed word forms,
  • source vendor: Daniel Naber
  • LICENSE: Creative Commons Attribution-ShareAlike 4.0 International (CC BY-SA 4.0)

• French (tagset reference)

  • data for assigning lemmata and tags to analyzed word forms,
  • source vendor: Lefff (Lexique des Formes Fléchies du Français / Lexicon of French inflected forms)
  • LICENSE: LGPL-LR free software license

• Italian (tagset reference)

  • data for assigning lemmata and tags to analyzed word forms,
  • source vendor: MORPH-IT
  • LICENSE: Creative Commons Attribution-ShareAlike 3.0 Unported License and GNU Lesser General Public License, version 2.1

• English (tagset reference)

  • data for assigning lemmata and tags to analyzed word forms,
  • source vendor: FreeLing
  • LICENSE: WordNet 3.0 license

• Portuguese (tagset reference)

  • data for assigning lemmata and tags to analyzed word forms,
  • source vendor: FreeLing
  • LICENSE: Creative Commons Attribution-ShareAlike 3.0 Unported License

• Catalan (tagset reference)

  • data for assigning lemmata and tags to analyzed word forms,
  • source vendor: FreeLing
  • LICENSE: Creative Commons Attribution 3.0 Unported License

• Welsh (tagset reference)

  • data for assigning lemmata and tags to analyzed word forms,
  • source vendor: FreeLing
  • LICENSE: GNU General Public License

• Spanish (tagset reference)

  • data for assigning lemmata and tags to analyzed word forms,
  • source vendor: FreeLing
  • LICENSE: Creative Commons Attribution 3.0 Unported License

• Galician (tagset reference)

  • data for assigning lemmata and tags to analyzed word forms,
  • source vendor: FreeLing
  • LICENSE: Creative Commons Attribution 3.0 Unported License

• Asturian (tagset reference)

  • data for assigning lemmata and tags to analyzed word forms,
  • source vendor: developed at the Universidad the Oviedo in the framework of the Eslema project
  • LICENSE: GNU General Public License

• Russian (tagset reference)

  • data for assigning lemmata and tags to analyzed word forms,
  • source vendor: FreeLing
  • LICENSE: GNU General Public License

Contact

Pavel Šmerk, Ph.D.
ma@nlp.fi.muni.cz
Natural Language Processing Centre
Faculty of Informatics, Masaryk University
Botanická 68a, 602 00, Brno, Czech Republic