Model adaptation for statistical machine translation

Author: 
Peer reviewed: 
No, item is not peer reviewed.
Date created: 
2009
Keywords: 
Natural language processing
Statistical machine translation
Back translations
Domain adaptation
Model adaptation
Unsupervised MERT
Abstract: 

Statistical machine translation (SMT) systems use statistical learning methods to learn how to translate from large amounts of parallel training data. Unfortunately, SMT systems are tuned to the domain of the training data and need to be adapted before they can be used to translate data in a different domain. First, we consider a semi-supervised technique to perform model adaptation. We explore new feature extraction techniques, feature combinations and their effects on performance. In addition, we introduce an unsupervised variant of Minimum Error Rate Training (MERT), which can be used to tune the SMT model parameters. We do this by using another SMT model that translates in the reverse direction. We apply this variant of MERT to the model adaptation task. Both of the techniques we explore in this thesis produce promising results in exhaustive experiments we performed for translation from French to English in different domains.

Language: 
English
Document type: 
Thesis
Rights: 
Copyright remains with the author. The author granted permission for the file to be printed, but not for the text to be copied and pasted.
File(s): 
Senior supervisor: 
A
Department: 
School of Computing Science - Simon Fraser University
Thesis type: 
(Computing Science) Thesis (M.Sc.)
Statistics: