Show simple item record

dc.contributor.authorGupta, Rohit
dc.contributor.authorOrasan, Constantin
dc.contributor.authorZampieri, Marcos
dc.contributor.authorVela, Mihaela
dc.contributor.authorvan Genabith, Josef
dc.contributor.authorMitkov, Ruslan
dc.date.accessioned2016-11-09T16:12:24Z
dc.date.available2016-11-09T16:12:24Z
dc.date.issued2016-11-02
dc.identifier.citationGupta, R., Orăsan, C., Zampieri, M. et al. (2016) Improving translation memory matching and retrieval using paraphrases. Machine Translation 30 (1), pp 19–40.en
dc.identifier.issn0922-6567en
dc.identifier.doi10.1007/s10590-016-9180-0
dc.identifier.urihttp://hdl.handle.net/2436/620274
dc.descriptionThis is an accepted manuscript of an article published by Springer Nature in Machine Translation on 02/11/2016, available online: https://doi.org/10.1007/s10590-016-9180-0 The accepted version of the publication may differ from the final published version.
dc.description.abstractMost of the current Translation Memory (TM) systems work on string level (character or word level) and lack semantic knowledge while matching. They use simple edit-distance calculated on surface-form or some variation on it (stem, lemma), which does not take into consideration any semantic aspects in matching. This paper presents a novel and efficient approach to incorporating semantic information in the form of paraphrasing in the edit-distance metric. The approach computes edit-distance while efficiently considering paraphrases using dynamic programming and greedy approximation. In addition to using automatic evaluation metrics like BLEU and METEOR, we have carried out an extensive human evaluation in which we measured post-editing time, keystrokes, HTER, HMETEOR, and carried out three rounds of subjective evaluations. Our results show that paraphrasing substantially improves TM matching and retrieval, resulting in translation performance increases when translators use paraphrase-enhanced TMs.
dc.formatapplication/pdf
dc.language.isoenen
dc.publisherSpringer Natureen
dc.relation.urlhttp://link.springer.com/10.1007/s10590-016-9180-0en
dc.rightsArchived with thanks to Machine Translationen
dc.rights.urihttp://creativecommons.org/licenses/by-nc-nd/4.0/*
dc.subjectTranslation memory (TM)en
dc.subjectParaphrasingen
dc.subjectComputer aided translation (CAT)en
dc.subjectEdit distanceen
dc.subjectDynamic programmingen
dc.subjectGreedy approximationen
dc.titleImproving translation memory matching and retrieval using paraphrasesen
dc.typeJournal article
dc.identifier.journalMachine Translationen
dc.date.accepted2016-10-03
rioxxterms.funderUniversity of Wolverhamptonen
rioxxterms.identifier.projectFP7 ITN People's Programme #317471en
rioxxterms.versionAMen
rioxxterms.licenseref.urihttps://creativecommons.org/CC BY-NC-ND 4.0en
rioxxterms.licenseref.startdate2017-11-01en
dc.source.volume30
dc.source.issue1
dc.source.beginpage19
dc.source.endpage40
refterms.dateFCD2019-03-20T10:24:11Z
refterms.versionFCDAM
refterms.dateFOA2017-11-01T00:00:00Z
html.description.abstractMost of the current Translation Memory (TM) systems work on string level (character or word level) and lack semantic knowledge while matching. They use simple edit-distance calculated on surface-form or some variation on it (stem, lemma), which does not take into consideration any semantic aspects in matching. This paper presents a novel and efficient approach to incorporating semantic information in the form of paraphrasing in the edit-distance metric. The approach computes edit-distance while efficiently considering paraphrases using dynamic programming and greedy approximation. In addition to using automatic evaluation metrics like BLEU and METEOR, we have carried out an extensive human evaluation in which we measured post-editing time, keystrokes, HTER, HMETEOR, and carried out three rounds of subjective evaluations. Our results show that paraphrasing substantially improves TM matching and retrieval, resulting in translation performance increases when translators use paraphrase-enhanced TMs.


Files in this item

Thumbnail
Name:
TMwithPP15Sep16.pdf
Size:
233.3Kb
Format:
PDF

This item appears in the following Collection(s)

Show simple item record

Archived with thanks to Machine Translation
Except where otherwise noted, this item's license is described as Archived with thanks to Machine Translation