Please use this identifier to cite or link to this item: http://hdl.handle.net/1893/23459
Full metadata record
DC FieldValueLanguage
dc.contributor.authorWang, Longyue-
dc.contributor.authorTu, Zhaopeng-
dc.contributor.authorZhang, Xiaojun-
dc.contributor.authorLi, Hang-
dc.contributor.authorWay, Andy-
dc.contributor.authorLiu, Qun-
dc.date.accessioned2016-12-05T22:13:34Z-
dc.date.available2016-12-05T22:13:34Z-
dc.date.issued2016-06-01-
dc.identifier.urihttp://hdl.handle.net/1893/23459-
dc.description.abstractDropped Pronouns (DP) in which pronouns are frequently dropped in the source language but should be retained in the target language are challenge in machine translation. In response to this problem, we propose a semisupervised approach to recall possibly missing pronouns in the translation. Firstly, we build training data for DP generation in which the DPs are automatically labelled according to the alignment information from a parallel corpus. Secondly, we build a deep learning-based DP generator for input sentences in decoding when no corresponding references exist. More specifically, the generation is two-phase: (1) DP position detection, which is modeled as a sequential labelling task with recurrent neural networks; and (2) DP prediction, which employs a multilayer perceptron with rich features. Finally, we integrate the above outputs into our translation system to recall missing pronouns by both extracting rules from the DP-labelled training data and translating the DP-generated input sentences. Experimental results show that our approach achieves a significant improvement of 1.58 BLEU points in translation performance with 66% F-score for DP generation accuracy.en_UK
dc.language.isoen-
dc.publisherThe Association for Computational Linguistics-
dc.relationWang L, Tu Z, Zhang X, Li H, Way A & Liu Q (2016) A Novel Approach to Dropped Pronoun Translation In: Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Stroudsburg, PA, USA: The Association for Computational Linguistics. The 15th Annual Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL HLT 2016), 12.6.2016 - 17.6.2016, San Diego, USA, pp. 983-993.-
dc.rightsThis paper is licensed under a Creative Commons Attribution 4.0 License.-
dc.titleA Novel Approach to Dropped Pronoun Translationen_UK
dc.typeConference Paperen_UK
dc.citation.spage983-
dc.citation.epage993-
dc.citation.publicationstatusPublished-
dc.citation.peerreviewedRefereed-
dc.type.statusBook Chapter: publisher version-
dc.identifier.urlhttp://aclweb.org/anthology/N/N16/N16-1113.pdf-
dc.author.emailxiaojun.zhang@stir.ac.uk-
dc.citation.btitleProceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies-
dc.citation.conferencedates2016-06-12T00:00:00Z-
dc.citation.conferencelocationSan Diego, USA-
dc.citation.conferencenameThe 15th Annual Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL HLT 2016)-
dc.citation.date01/06/2016-
dc.citation.isbn978-1-941643-91-4-
dc.publisher.addressStroudsburg, PA, USA-
dc.contributor.affiliationDublin City University-
dc.contributor.affiliationHuawei Technologies (HK)-
dc.contributor.affiliationEnglish Studies-
dc.contributor.affiliationHuawei Technologies (HK)-
dc.contributor.affiliationADAPT Centre-
dc.contributor.affiliationADAPT Centre-
Appears in Collections:Literature and Languages Conference Papers and Proceedings

Files in This Item:
File Description SizeFormat 
N16-1113.pdf593.76 kBAdobe PDFView/Open


This item is protected by original copyright



Items in the Repository are protected by copyright, with all rights reserved, unless otherwise indicated.

If you believe that any material held in STORRE infringes copyright, please contact library@stir.ac.uk providing details and we will remove the Work from public display in STORRE and investigate your claim.