El jue, 09-04-2009 a las 09:13 +0200, Francis Tyers escribió:
> Dear corpora-list members,
> I would be interested in receiving information/references about any
> articles/reports/papers/etc. on the subject of machine translation for
> Celtic languages. The ones I am aware of are:
> * cy-en (2001) John D. Phillips: "The Bible as a Basis for Machine
> Translation". PACL.
> * kw-en (2003) Paul R. Bowden "Building a Lexicon for a Kernewek MT
> System" TALN
> * cy-en (2004) Harold Somers: "Machine Translation and Welsh: The Way
> Forward; Cyfieithu Peirianyddol a?r Gymraeg: Y Ffordd Ymlaen". Report
> for the Welsh Language Board.
> * ga-gd (2006) Kevin P. Scannell: "Machine translation for closely
> related language pairs". LREC-2006. pp.103-107.
> * cy-en (2006) Dafydd Jones & Andreas Eisele: "Phrase-based statistical
> machine translation between English and Welsh". LREC-2006. pp.75-77.
> [PDF, 54KB]
> * cy-en (2009) Francis M. Tyers & Kevin Donnelly: "apertium-cy: a
> collaboratively-developed RBMT system for Welsh to English". PBML 91,
> 2009; pp.57-66.
Aside from the above papers, I received one further reference to machine translation and Celtic languages,
* David Talbot and Miles Osborne. Modelling Lexical Redundancy for Machine Translation. ACL, Sydney, Australia 2006. http://www.iccs.inf.ed.ac.uk/~osborne/papers/acl06.pdf
NB. There is an errata. The Welsh corpus they use is _not_ in the public domain. It can be downloaded from.
Which describes describes a technique that learns to collapse lexical distinctions that don't matter for translation into the target language. They apply it for several languages, including Welsh.
Many thanks to those who replied,