[ 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 2011
Type in proceedings
Status published
Language English
Author(s) Macháček, Matouš Bojar, Ondřej
Title Approximating a Deep-Syntactic Metric for MT Evaluation and Tuning
Czech title Aproximace hloubkově-syntaktické metriky pro vyhodnocování a optimalizaci strojového překladu
Proceedings 2011: Edinburgh, UK: WMT 2011 (EMNLP): Proceedings of the Sixth Workshop on Statistical Machine Translation
Pages range 92-98
URL http://www.aclweb.org/anthology/W11-2108
Supported by 2009-2012 FP7-ICT-2007-3-231720 (EuroMatrix Plus) 2009-2012 7E09003 (EuroMatrixPlus – Bringing Machine Translation for European Languages to the User) 2010-2012 GPP406/10/P259 (Hybridní frázový a hloubkově-syntaktický strojový překlad) 2011-2013 GAP406/11/1499 (Čeština ve věku strojového překladu) 2005-2010 MSM 0021620838 (Moderní metody, struktury a systémy informatiky)
Czech abstract SemPOS je automatická metrika kvality strojového překladu založená na hloubkové syntaxi. Poměrně dobře koreluje s lidským hodnocením, ale je výpočetně drahá. V příspěvku zkoumáme její odlehčené verze a zapojujeme je i do iterativní optimalizace překladového modelu.
English abstract SemPOS is an automatic metric of machine translation quality for Czech and English focused on content words. It correlates well with human judgments but it is computationally costly and hard to adapt to other languages because it relies on a deep-syntactic analysis of the system output and the reference. To remedy this, we attempt at approximating SemPOS using only tagger output and a few heuristics. At a little expense in correlation to human judgments, we can evaluate MT systems much faster. Additionally, we describe our submission to the Tunable Metrics Task in WMT11.
Specialization linguistics ("jazykověda")
Confidentiality default – not confidential
Open access no
Editor(s)* Chris Callison-Burch; Philipp Koehn; Christof Monz; Omar F. Zaidan
ISBN* 978-1-937284-12-1
Address* Edinburgh, UK
Month* July
Venue* Informatics Forum, Edinburgh
Publisher* Association for Computational Linguistics
Institution* University of Edinburgh
Creator: Common Account
Created: 8/31/11 11:05 AM
Modifier: Almighty Admin
Modified: 3/5/12 4:19 PM
***

PaperpublicW11-2108.pdfapplication/pdf
posterpublic2011-FILE-machavcek-bojar_2011_WMT-CAMERA-READY.pd...application/pdf
posterpublic2011-FILE-machavcek-bojar_2011_WMT-2011-machacek-b...application/vnd.ms-powerpoint
Content, Design & Functionality: ÚFAL, 2006–2016. Page generated: Tue Oct 23 06:30:24 CEST 2018

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

100% OpenAIRE compliant