{"title":"Improving Machine Translation of English Relative Clauses with Automatic Text Simplification","authors":"Sanja Štajner, Maja Popovic","doi":"10.18653/v1/W18-7006","DOIUrl":null,"url":null,"abstract":"This article explores the use of automatic sentence simplification as a preprocessing step in neural machine translation of English relative clauses into grammatically complex languages. Our experiments on English-to-Serbian and English to-German translation show that this approach can reduce technical post-editing \neffort (number of post-edit operations) to \nobtain correct translation. We find that \nlarger improvements can be achieved for \nmore complex target languages, as well as \nfor MT systems with lower overall performance. The improvements mainly originate from correctly simplified sentences \nwith relatively complex structure, while \nsimpler structures are already translated \nsufficiently well using the original source \nsentences.","PeriodicalId":247947,"journal":{"name":"Proceedings of the 1st Workshop on Automatic Text Adaptation (ATA)","volume":"44 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"12","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the 1st Workshop on Automatic Text Adaptation (ATA)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.18653/v1/W18-7006","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 12
Abstract
This article explores the use of automatic sentence simplification as a preprocessing step in neural machine translation of English relative clauses into grammatically complex languages. Our experiments on English-to-Serbian and English to-German translation show that this approach can reduce technical post-editing
effort (number of post-edit operations) to
obtain correct translation. We find that
larger improvements can be achieved for
more complex target languages, as well as
for MT systems with lower overall performance. The improvements mainly originate from correctly simplified sentences
with relatively complex structure, while
simpler structures are already translated
sufficiently well using the original source
sentences.