Please use this identifier to cite or link to this item: https://doi.org/10.21256/zhaw-20320
Full metadata record
DC FieldValueLanguage
dc.contributor.authorCampos, Jon Ander-
dc.contributor.authorOtegi, Arantxa-
dc.contributor.authorSoroa, Aitor-
dc.contributor.authorDeriu, Jan Milan-
dc.contributor.authorCieliebak, Mark-
dc.contributor.authorAgirre, Eneko-
dc.date.accessioned2020-08-05T15:22:33Z-
dc.date.available2020-08-05T15:22:33Z-
dc.date.issued2020-
dc.identifier.urihttps://digitalcollection.zhaw.ch/handle/11475/20320-
dc.description.abstractThe goal of this work is to build conversational Question Answering (QA) interfaces for the large body of domain-specific information available in FAQ sites. We present DoQA, a dataset with 2,437 dialogues and 10,917 QA pairs. The dialogues are collected from three Stack Exchange sites using the Wizard of Oz method with crowdsourcing. Compared to previous work, DoQA comprises well-defined information needs, leading to more coherent and natural conversations with less factoid questions and is multi-domain. In addition, we introduce a more realistic information retrieval (IR) scenario where the system needs to find the answer in any of the FAQ documents. The results of an existing, strong, system show that, thanks to transfer learning from a Wikipedia QA dataset and fine tuning on a single FAQ domain, it is possible to build high quality conversational QA systems for FAQs without in-domain training data. The good results carry over into the more challenging IR scenario. In both cases, there is still ample room for improvement, as indicated by the higher human upperbound.de_CH
dc.language.isoende_CH
dc.publisherAssociation for Computational Linguisticsde_CH
dc.rightshttp://creativecommons.org/licenses/by/4.0/de_CH
dc.subjectQuestion answeringde_CH
dc.subjectDeep learningde_CH
dc.subjectNatural language processingde_CH
dc.subject.ddc004: Informatikde_CH
dc.subject.ddc400: Sprache und Linguistikde_CH
dc.titleDoQA : accessing domain-specific FAQs via conversational QAde_CH
dc.typeKonferenz: Paperde_CH
dcterms.typeTextde_CH
zhaw.departementSchool of Engineeringde_CH
zhaw.organisationalunitInstitut für Angewandte Informationstechnologie (InIT)de_CH
dc.identifier.doi10.18653/v1/2020.acl-main.652de_CH
dc.identifier.doi10.21256/zhaw-20320-
zhaw.conference.detailsACL 2020, Virtual, 5-10 July 2020de_CH
zhaw.funding.euNode_CH
zhaw.originated.zhawYesde_CH
zhaw.pages.end7314de_CH
zhaw.pages.start7302de_CH
zhaw.publication.statuspublishedVersionde_CH
zhaw.publication.reviewPeer review (Publikation)de_CH
zhaw.title.proceedingsProceedings of the 58th Annual Meeting of the Association for Computational Linguisticsde_CH
zhaw.webfeedSoftware Systemsde_CH
zhaw.funding.zhawLIHLITH - Learning to Interact with Humans by Lifelong Interaction with Humansde_CH
zhaw.author.additionalNode_CH
zhaw.display.portraitYesde_CH
Appears in collections:Publikationen School of Engineering

Files in This Item:
File Description SizeFormat 
2020_Campos-etal_DoQA_ACL.pdf590.12 kBAdobe PDFThumbnail
View/Open


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