{"id":190490,"updated":"2025-01-20T01:11:51.499044+00:00","links":{},"created":"2025-01-19T00:56:27.234388+00:00","metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00190490","sets":["581:9322:9329"]},"path":["9329"],"owner":"11","recid":"190490","title":["A Comprehensive Empirical Comparison of Domain Adaptation Methods for Neural Machine Translation"],"pubdate":{"attribute_name":"公開日","attribute_value":"2018-07-15"},"_buckets":{"deposit":"68828ed3-8493-42f5-a99d-2b2869ca7726"},"_deposit":{"id":"190490","pid":{"type":"depid","value":"190490","revision_id":0},"owners":[11],"status":"published","created_by":11},"item_title":"A Comprehensive Empirical Comparison of Domain Adaptation Methods for Neural Machine Translation","author_link":["435920","435923","435919","435924","435922","435921"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"A Comprehensive Empirical Comparison of Domain Adaptation Methods for Neural Machine Translation"},{"subitem_title":"A Comprehensive Empirical Comparison of Domain Adaptation Methods for Neural Machine Translation","subitem_title_language":"en"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"[一般論文] neural machine translation, domain adaptation, empirical comparison","subitem_subject_scheme":"Other"}]},"item_type_id":"2","publish_date":"2018-07-15","item_2_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"Institute for Datability Science, Osaka University"},{"subitem_text_value":"Graduate School of Informatics, Kyoto University"},{"subitem_text_value":"Graduate School of Informatics, Kyoto University"}]},"item_2_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Institute for Datability Science, Osaka University","subitem_text_language":"en"},{"subitem_text_value":"Graduate School of Informatics, Kyoto University","subitem_text_language":"en"},{"subitem_text_value":"Graduate School of Informatics, Kyoto University","subitem_text_language":"en"}]},"item_language":{"attribute_name":"言語","attribute_value_mlt":[{"subitem_language":"eng"}]},"publish_status":"0","weko_shared_id":-1,"item_file_price":{"attribute_name":"Billing file","attribute_type":"file","attribute_value_mlt":[{"url":{"url":"https://ipsj.ixsq.nii.ac.jp/record/190490/files/IPSJ-JNL5907005.pdf","label":"IPSJ-JNL5907005.pdf"},"date":[{"dateType":"Available","dateValue":"2020-07-15"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-JNL5907005.pdf","filesize":[{"value":"779.2 kB"}],"mimetype":"application/pdf","priceinfo":[{"tax":["include_tax"],"price":"0","billingrole":"5"},{"tax":["include_tax"],"price":"0","billingrole":"6"},{"tax":["include_tax"],"price":"0","billingrole":"8"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"74aa4932-70d2-4c2f-9ef9-6e6982401f87","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2018 by the Information Processing Society of Japan"}]},"item_2_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Chenhui, Chu"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Raj, Dabre"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Sadao, Kurohashi"}],"nameIdentifiers":[{}]}]},"item_2_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Chenhui, Chu","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Raj, Dabre","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Sadao, Kurohashi","creatorNameLang":"en"}],"nameIdentifiers":[{}]}]},"item_2_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AN00116647","subitem_source_identifier_type":"NCID"}]},"item_resource_type":{"attribute_name":"資源タイプ","attribute_value_mlt":[{"resourceuri":"http://purl.org/coar/resource_type/c_6501","resourcetype":"journal article"}]},"item_2_source_id_11":{"attribute_name":"ISSN","attribute_value_mlt":[{"subitem_source_identifier":"1882-7764","subitem_source_identifier_type":"ISSN"}]},"item_2_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"Neural machine translation (NMT) has shown very promising results when there are large amounts of parallel corpora. However, for low resource domains, vanilla NMT cannot give satisfactory performance due to overfitting on the small size of parallel corpora. Two categories of domain adaptation approaches have been proposed for low resource NMT, i.e., adaptation using out-of-domain parallel corpora and in-domain monolingual corpora. In this paper, we conduct a comprehensive empirical comparison of the methods in both categories. For domain adaptation using out-of-domain parallel corpora, we further propose a novel domain adaptation method named mixed fine tuning, which combines two existing methods namely fine tuning and multi domain NMT. For domain adaptation using in-domain monolingual corpora, we compare two existing methods namely language model fusion and synthetic data generation. In addition, we propose a method that combines these two categories. We empirically compare all the methods and discuss their benefits and shortcomings. To the best of our knowledge, this is the first work on a comprehensive empirical comparison of domain adaptation methods for NMT.\n------------------------------\nThis is a preprint of an article intended for publication Journal of\nInformation Processing(JIP). This preprint should not be cited. This\narticle should be cited as: Journal of Information Processing Vol.26(2018) (online)\nDOI　http://dx.doi.org/10.2197/ipsjjip.26.529\n------------------------------","subitem_description_type":"Other"}]},"item_2_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"Neural machine translation (NMT) has shown very promising results when there are large amounts of parallel corpora. However, for low resource domains, vanilla NMT cannot give satisfactory performance due to overfitting on the small size of parallel corpora. Two categories of domain adaptation approaches have been proposed for low resource NMT, i.e., adaptation using out-of-domain parallel corpora and in-domain monolingual corpora. In this paper, we conduct a comprehensive empirical comparison of the methods in both categories. For domain adaptation using out-of-domain parallel corpora, we further propose a novel domain adaptation method named mixed fine tuning, which combines two existing methods namely fine tuning and multi domain NMT. For domain adaptation using in-domain monolingual corpora, we compare two existing methods namely language model fusion and synthetic data generation. In addition, we propose a method that combines these two categories. We empirically compare all the methods and discuss their benefits and shortcomings. To the best of our knowledge, this is the first work on a comprehensive empirical comparison of domain adaptation methods for NMT.\n------------------------------\nThis is a preprint of an article intended for publication Journal of\nInformation Processing(JIP). This preprint should not be cited. This\narticle should be cited as: Journal of Information Processing Vol.26(2018) (online)\nDOI　http://dx.doi.org/10.2197/ipsjjip.26.529\n------------------------------","subitem_description_type":"Other"}]},"item_2_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographic_titles":[{"bibliographic_title":"情報処理学会論文誌"}],"bibliographicIssueDates":{"bibliographicIssueDate":"2018-07-15","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"7","bibliographicVolumeNumber":"59"}]},"relation_version_is_last":true,"weko_creator_id":"11"}}