Arabic

From Apertium
Jump to navigation Jump to search

Arabic is a semitic language (http://en.wikipedia.org/wiki/Hamito-Semitic). There is currenly only one pair in development:

  1. staging mt-ar: Maltese-Arabic. https://svn.code.sf.net/p/apertium/svn/staging/apertium-mt-ar/

Developping other semitic languages pair with Arabic would be a good idea. Example: Tamazight, Hebrew.

See also

Resources

  • Sarf - Arabic Morphology System (all in Java...)
  • ElixirFM (online interface here) is a Functional Arabic Morphology written in Haskell and Perl; the lexicon is a "re-processed" version of the Buckwalter analyser.
  • There is a good documentation of how to make a morphological analyser for Arabic (and Semitic languages in general) in the Beesley/Karttunen finite state transducer book, documenting the Xerox compiler (Ken Beesley also made an Arabic fst). Also, there now is an open source compiler reading the Xerox format, the HFST compiler.
  • And there is also an open source finite state morphological analyser for Arabic, AraComLex (online interface here). Among other resources related to AraComLex there is a list of Arabic morphological patterns and a frequency word list from a 1 billion word corpus.

Corpora