9th Annual Conference of the International Speech Communication Association

Brisbane, Australia
September 22-26, 2008

Discovering Phrases in Machine Translation by Simulated Annealing

Caroline Lavecchia, David Langlois, Kamel Smaïli

LORIA, France

In this paper, we propose a new phrase-based translation model based on inter-lingual triggers. The originality of our method is double. First we identify common source. Then we use inter-lingual triggers in order to retrieve their translations. Furthermore, we consider the way of extracting phrase translations as an optimization issue. For that we use simulated annealing algorithm to find out the best phrase translations among all those determined by inter-lingual triggers. The best phrases are those which improve the translation quality in terms of Bleu score. Tests are achieved on the proceedings of the European Parliament corpora. The training is made on a corpus containing 596K parallel sentences (French-English) and tests on a corpus of 1444 sentences. With only 8.1% of the identified source phrases occurring in the test corpus, our system overcomes the baseline model by almost 3 points.

Full Paper

Bibliographic reference.  Lavecchia, Caroline / Langlois, David / Smaïli, Kamel (2008): "Discovering phrases in machine translation by simulated annealing", In INTERSPEECH-2008, 2354-2357.