Nano Gough Postgraduate Research Student School of Computing Dublin City University Dublin 9 Ireland B.Sc in Computational Linguistics Degree sought PhD in Computing Year 3 Supervisor Dr. Andy Way Contact Details Ext +353-1-7005618 ngough@computing.dcu.ie

Example Based Machine Translation via the Web Example-Based Machine Translation (EBMT) involves the use of previously translated examples to produce new translations. My approach is a linguistics-lite one. The only linguistic information which I currently apply is that of the 'Marker Hypothesis' (Green, 1979). The Marker Hypothesis is a universal psycholinguistic constraint, which states that natural languages are 'marked' for complex syntactic structure at surface form by a closed set of specific lexemes and morphemes. That is, a closed list of determiners, prepositions, quantifiers, pronouns and conjuncts can be used to 'mark' source and target phrases at distinct points, signifying the beginning and end of smaller fragments. I train my EBMT system, using lexical resources deduced from the Marker Hypothesis. I use the WWW to do automatic post-hoc validation of the translations produced by my system. My PhD research is funded by an IBM fellowship. Publications Gough, N., A. Way. and M. Hearne. 2002. 'Example-Based Machine Translation via the Web'. In S. Richardson (ed.) Machine Translation: From Research to Real Users, LNAI 2499, In Proceedings of 5th Conference of the Association for Machine Translation in the Americas, AMTA 2002, Tiburon, CA, pp.74--83. A. Way & N. Gough. 2003. Developing and Validating an Example-Based Machine Translation System using the World Wide Web, Computational Linguistics: special issue on Web as Corpus (2003) Gough, N. and A. Way. Controlled Generation in example based Machine Translation. In MT Summit IX (2003) New Orleans, LA., pp. 133-140. Gough, N. and A. Way. 2003. Example based Controlled Translation. In Proceedings of the 9th Workshop of the EAMT , Malta, [forthcoming]. Teaching I have taught module LC233 (Introduction to Programming in Perl) to 2nd year students on the BSc. in Computational Linguistics.