Can Out-of-the-box NMT Beat a Domain-trained Moses on Technical Data

Anne Beyer, Vivien Macketanz, Aljoscha Burchardt, Philip Williams

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In the last year, we have seen a lot of evidence about the superiority of neural
machine translation approaches (NMT) over phrase-based statistical approaches
(PBMT). This trend has shown for the general domain at public competitions such as the WMT challenges as well as in the obvious quality increase in online translation services that have changed their technology. In this paper, we take the perspective of an LSP. The questions we want to answer with this study is if now is already the time to invest in the new technology. To answer this question, we have collected evidence as to whether an existing state of-the-art NMT system for the general domain can already compete with a domain trained and optimised Moses (PBMT) system or if it is maybe already better. As it is well known that automatic quality measures are not reliable for comparing the performance of different system types, we have performed a detailed manual evaluation based on a test suite of domain segments.
Original languageEnglish
Title of host publicationEAMT 2017: The 20th Annual Conference of the European Association for Machine Translation
Number of pages6
Publication statusPublished - 31 May 2017
Event20th Annual Conference of the European Association for Machine Translation - Prague, Czech Republic
Duration: 29 May 201731 May 2017
https://ufal.mff.cuni.cz/eamt2017/

Conference

Conference20th Annual Conference of the European Association for Machine Translation
Abbreviated titleEAMT 2017
Country/TerritoryCzech Republic
CityPrague
Period29/05/1731/05/17
Internet address

Cite this