[ Skip to the content ]

Institute of Formal and Applied Linguistics

at Faculty of Mathematics and Physics, Charles University, Prague, Czech Republic


[ Back to the navigation ]

Publication


Year 2010
Type oral presentation *
Status published
Language English
Author(s) Mareček, David
Title Maximum Entropy Translation Model in Dependency-Based MT Framework
Czech title Maximum Entropy překladový model v rámci závislostního strojového překladu
Publisher's city and country Uppsala, Sweeden
Venue Uppsala University
Month July
URL http://ufal.mff.cuni.cz/pire10/
Supported by 2010-2012 GAUK 116310/2010 (Anglicko-český strojový překlad s využitím hloubkové syntaxe)
English abstract We propose a forward translation model consisting of a set of maximum entropy classifiers: a separate classifier is trained for each (sufficiently frequent) source-side lemma. In this way the estimates of translation probabilities can be sensitive to a large number of features derived from the source sentence (including non-local features, features making use of sentence syntactic structure, etc.). When integrated into English-to-Czech dependency-based translation scenario implemented in the TectoMT framework, the new translation model significantly outperforms the baseline model (MLE) in terms of BLEU. The performance is further boosted in a configuration inspired by Hidden Tree Markov Models which combines the maximum entropy translation model with the target-language dependency tree model.
Specialization linguistics ("jazykověda")
Confidentiality default – not confidential
Event PIRE meeting
Presentation type other
Open access no
Creator: Common Account
Created: 10/30/10 1:32 PM
Modifier: Common Account
Modified: 11/12/10 9:23 PM
***

Content, Design & Functionality: ÚFAL, 2006–2016. Page generated: Mon Oct 15 18:34:17 CEST 2018

[ Back to the navigation ] [ Back to the content ]

100% OpenAIRE compliant