A Syntax-based Translation Model

Kenji Yamada

Xerox Research Centre Europe

Le 17 mars 2004 à 11 h 30

Salle 3195, Pavillon André-Aisenstadt


I present a syntax-based statistical translation model. Unlike word-based IBM translation models or other phrase-based models, the model transforms a source-language parse tree into a target-language string by applying stochastic operations at each node. By using a parse tree, the model can more easily capture linguistic differences such as word order and case marking. Model parameters are estimated in polynomial time using an EM algorithm. A decoding algorithm for the model will be also presented.


Pour recevoir les annonces hebdomadaires par courriel, visitez http://rali.iro.umontreal.ca/rali/?q=fr/node/1631

Liste des autres séminaires