Cascaded model adaptation for dialog act segmentation and tagging

Basit Öğe Kaydı
dc.authorid0000-0002-4597-0954
dc.authorid0000-0001-5246-2117
dc.contributor.authorGüz, Ümiten_US
dc.contributor.authorTür, Gökhanen_US
dc.contributor.authorHakkani Tür, Dileken_US
dc.contributor.authorCuendet, Sebastienen_US
dc.date.accessioned2015-01-15T23:01:36Z
dc.date.available2015-01-15T23:01:36Z
dc.date.issued2010-04
dc.departmentIşık Üniversitesi, Mühendislik Fakültesi, Elektrik-Elektronik Mühendisliği Bölümüen_US
dc.departmentIşık University, Faculty of Engineering, Department of Electrical-Electronics Engineeringen_US
dc.descriptionThis material is based upon work supported by Defense Advanced Research Projects Agency (DARPA) CALO (Contract No. FA8750-07-D-0185, Delivery Order 0004), the Scientific and Technological Research Council of Turkey (TUBITAK) fundings at SRI, Isik University Research Fund (Contract No. 0513304), J. William Fulbright Post-Doctoral Research Fellowship, and the Swiss National Science Foundation through the research network, IM2 fundings at ICS1. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the funding agencies. We thank Elizabeth Shriberg, Andreas Stoleke, Matthias Zimmerman, and Matthew Magimai Doss for many helpful discussionsen_US
dc.description.abstractThere are many speech and language processing problems which require cascaded classification tasks. While model adaptation has been shown to be useful in isolated speech and language processing tasks, it is not clear what constitutes system adaptation for such complex systems. This paper studies the following questions: In cases where a sequence of classification tasks is employed, how important is to adapt the earlier or latter systems? Is the performance improvement obtained in the earlier stages via adaptation carried on to later stages in cases where the later stages perform adaptation using similar data and/or methods? In this study, as part of a larger scale multiparty meeting understanding system, we analyze various methods for adapting dialog act segmentation and tagging models trained on conversational telephone speech (CTS) to meeting style conversations. We investigate the effect of using adapted and unadapted models for dialog act segmentation with those of tagging, showing the effect of model adaptation for cascaded classification tasks. Our results indicate that we can achieve significantly better dialog act segmentation and tagging by adapting the out-of-domain models, especially when the amount of in-domain data is limited. Experimental results show that it is more effective to adapt the models in the latter classification tasks, in our case dialog act tagging, when dealing with a sequence of cascaded classification tasksen_US
dc.description.sponsorshipCALOen_US
dc.description.sponsorshipIsik Universityen_US
dc.description.sponsorshipTUBITAKen_US
dc.description.sponsorshipDefense Advanced Research Projects Agencyen_US
dc.description.sponsorshipJ. William Fulbright College of Arts and Sciences, University of Arkansasen_US
dc.description.sponsorshipSchweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschungen_US
dc.description.versionPublisher's Versionen_US
dc.identifier.citationGüz, Ü., Tür, G., Hakkani Tür, D. & Cuendet, S. (2010). Cascaded model adaptation for dialog act segmentation and tagging. Computer Speech and Language, 24(2), 289-306. doi:10.1016/j.csl.2009.04.006en_US
dc.identifier.doi10.1016/j.csl.2009.04.006
dc.identifier.endpage306
dc.identifier.issn0885-2308
dc.identifier.issn1095-8363
dc.identifier.issue2
dc.identifier.scopus2-s2.0-70349249525
dc.identifier.scopusqualityQ2
dc.identifier.startpage289
dc.identifier.urihttps://hdl.handle.net/11729/372
dc.identifier.urihttp://dx.doi.org/10.1016/j.csl.2009.04.006
dc.identifier.volume24
dc.identifier.wosWOS:000271445100010
dc.identifier.wosqualityQ2
dc.indekslendigikaynakWeb of Scienceen_US
dc.indekslendigikaynakScopusen_US
dc.indekslendigikaynakScience Citation Index Expanded (SCI-EXPANDED)en_US
dc.institutionauthorGüz, Ümiten_US
dc.institutionauthorid0000-0002-4597-0954
dc.language.isoenen_US
dc.peerreviewedYesen_US
dc.publicationstatusPublisheden_US
dc.publisherElsevier Ltden_US
dc.relation.ispartofComputer Speech and Languageen_US
dc.relation.publicationcategoryMakale - Uluslararası Hakemli Dergi - Kurum Öğretim Elemanıen_US
dc.rightsinfo:eu-repo/semantics/closedAccessen_US
dc.subjectModel adaptationen_US
dc.subjectDialog act segmentationen_US
dc.subjectDialog act taggingen_US
dc.subjectMeetings processingen_US
dc.subjectSpeechen_US
dc.subjectRecognitionen_US
dc.subjectSystemen_US
dc.titleCascaded model adaptation for dialog act segmentation and taggingen_US
dc.typeArticleen_US
dspace.entity.typePublication

Dosyalar

Orijinal paket
Listeleniyor 1 - 1 / 1
Küçük Resim Yok
İsim:
372.pdf
Boyut:
783.32 KB
Biçim:
Adobe Portable Document Format
Açıklama:
Publisher's Version
İndir

Koleksiyon

Makale Koleksiyonu | Elektrik-Elektronik Mühendisliği Bölümü
Scopus İndeksli Yayınlar Koleksiyonu
WoS İndeksli Yayınlar Koleksiyonu