Shallow-transfer machine translation engine

 An open-source shallow-transfer machine translation
 engine, Apertium is initially aimed at related-language pairs.
 .
 It uses finite-state transducers for lexical processing,
 hidden Markov models for part-of-speech tagging, and
 finite-state based chunking for structural transfer.
 .
 The system is largely based upon systems already developed by
 the Transducens group at the Universitat d'Alacant, such as
 interNOSTRUM (Spanish-Catalan, http://www.internostrum.com/welcome.php)
 and Traductor Universia (Spanish-Portuguese,
 http://traductor.universia.net).
 .
 It will be possible to use Apertium to build machine translation
 systems for a variety of related-language pairs simply providing
 the linguistic data needed in the right format.