[ 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 2016
Type book chapter/part
Status published
Language English
Author(s) Green, Nathan David Žabokrtský, Zdeněk
Title Creating Hybrid Dependency Parsers for Syntax-Based MT
Czech title Hybridní závislostní analyzátory pro strojový překlad založený na syntaxi
Book title Hybrid Approaches to Machine Translation
Publisher Springer International Publishing
Publisher's city and country Switzerland
Pages range 161-190
Total book pages 205
How published print
URL http://link.springer.com/book/10.1007%2F978-3-319-21311-8
Supported by 2016-2019 LM2015071 (LINDAT-CLARIN: Institut pro analýzu, zpracování a distribuci lingvistických dat) 2012-2016 PRVOUK P46 (Informatika)
ISBN 978-3-319-21310-1
Czech abstract Závislostní parsery jsou obvykle vyhodnocovány pomocí podílu správně zavěšených uzlů, toto skóre ale neříká nic o užitečnosti daného parseru pro koncové aplikace, v nichž slouží jako komponenta. V této kapitole se zabývá tím, jak rozdílné parsery a jejich různé kombinace přispívají ke kvalitě strojového překladu založeného na syntaxi. Překládáme výsledky pro několik základních typů parserů a pro různé metody jejich kombinace. Ukazujeme korelace se standardními metrikami pro vyhodnocování kvality strojového překladu.
English abstract Dependency parsers are almost ubiquitously evaluated on their accuracy scores, these scores say nothing of the complexity and usefulness of the resulting structures. As dependency parses are basic structures in which other systems are built upon, it would seem more reasonable to judge these parsers down the NLP pipeline. In this chapter, we will discuss how different forms and different hybrid combinations of dependency parses effect the overall output of Syntax-Based machine translation both through automatic and manual evaluation. We show results from a variety of individual parsers, including dependency and constituent parsers, and describe multiple ensemble parsing techniques with their overall effect on the Machine Translation system. We show that parsers’ UAS scores are more correlated to the NIST evaluation metric than to the BLEU Metric, however we see increases in both metrics. To truly see the effect of hybrid dependency parsers on machine translation, we will describe and evaluate a combined resource we have released, that contains gold standard dependency trees along with gold standard translations.
Specialization linguistics ("jazykověda")
Confidentiality default – not confidential
Book type monograph
Role of the author(s) chapter author(s)
Open access no
DOI 10.1007/978-3-319-21311-8
Creator: Common Account
Created: 10/23/16 11:46 AM
Modifier: Almighty Admin
Modified: 2/25/17 10:06 PM
***

Content, Design & Functionality: ÚFAL, 2006–2016. Page generated: Sat Sep 23 23:57:25 CEST 2017

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

100% OpenAIRE compliant