User Evaluation of a Multi-dimensional Statistical Dialogue System

Simon Keizer; Ondrej Dusek; Xingkun Liu; Verena Rieser

doi:10.18653/v1/W19-5945

User Evaluation of a Multi-dimensional Statistical Dialogue System

Simon Keizer, Ondrej Dusek, Xingkun Liu, Verena Rieser

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

50 Downloads (Pure)

Abstract

We present the first complete spoken dialogue system driven by a multi-dimensional statistical dialogue manager. This framework has been shown to substantially reduce data needs by leveraging domain-independent dimensions, such as social obligations or feedback, which (as we show) can be transferred between domains. In this paper, we conduct a user study and show that the performance of a multi-dimensional system, which can be adapted from a source domain, is equivalent to that of a one-dimensional baseline, which can only be trained from scratch.

Original language	English
Title of host publication	Proceedings of the 20th Annual SIGdial Meeting on Discourse and Dialogue
Publisher	Association for Computational Linguistics
Pages	392–398
Number of pages	7
ISBN (Electronic)	9781950737611
DOIs	https://doi.org/10.18653/v1/W19-5945
Publication status	Published - 2019
Event	20th Annual SIGdial Meeting on Discourse and Dialogue 2019 - Stockholm, Sweden Duration: 11 Sept 2019 → 13 Sept 2019

Conference

Conference	20th Annual SIGdial Meeting on Discourse and Dialogue 2019
Country/Territory	Sweden
City	Stockholm
Period	11/09/19 → 13/09/19

Access to Document

10.18653/v1/W19-5945Licence: CC BY

Download pdfFinal published version, 517 KBLicence: CC BY

Cite this

@inproceedings{55e986465ba54cc5a197f52a1cf20890,

title = "User Evaluation of a Multi-dimensional Statistical Dialogue System",

abstract = "We present the first complete spoken dialogue system driven by a multi-dimensional statistical dialogue manager. This framework has been shown to substantially reduce data needs by leveraging domain-independent dimensions, such as social obligations or feedback, which (as we show) can be transferred between domains. In this paper, we conduct a user study and show that the performance of a multi-dimensional system, which can be adapted from a source domain, is equivalent to that of a one-dimensional baseline, which can only be trained from scratch.",

author = "Simon Keizer and Ondrej Dusek and Xingkun Liu and Verena Rieser",

year = "2019",

doi = "10.18653/v1/W19-5945",

language = "English",

pages = "392–398",

booktitle = "Proceedings of the 20th Annual SIGdial Meeting on Discourse and Dialogue",

publisher = "Association for Computational Linguistics",

note = "20th Annual SIGdial Meeting on Discourse and Dialogue 2019 ; Conference date: 11-09-2019 Through 13-09-2019",

}

Keizer, S, Dusek, O, Liu, X & Rieser, V 2019, User Evaluation of a Multi-dimensional Statistical Dialogue System. in Proceedings of the 20th Annual SIGdial Meeting on Discourse and Dialogue. Association for Computational Linguistics, pp. 392–398, 20th Annual SIGdial Meeting on Discourse and Dialogue 2019, Stockholm, Sweden, 11/09/19. https://doi.org/10.18653/v1/W19-5945

TY - GEN

T1 - User Evaluation of a Multi-dimensional Statistical Dialogue System

AU - Keizer, Simon

AU - Dusek, Ondrej

AU - Liu, Xingkun

AU - Rieser, Verena

PY - 2019

Y1 - 2019

N2 - We present the first complete spoken dialogue system driven by a multi-dimensional statistical dialogue manager. This framework has been shown to substantially reduce data needs by leveraging domain-independent dimensions, such as social obligations or feedback, which (as we show) can be transferred between domains. In this paper, we conduct a user study and show that the performance of a multi-dimensional system, which can be adapted from a source domain, is equivalent to that of a one-dimensional baseline, which can only be trained from scratch.

AB - We present the first complete spoken dialogue system driven by a multi-dimensional statistical dialogue manager. This framework has been shown to substantially reduce data needs by leveraging domain-independent dimensions, such as social obligations or feedback, which (as we show) can be transferred between domains. In this paper, we conduct a user study and show that the performance of a multi-dimensional system, which can be adapted from a source domain, is equivalent to that of a one-dimensional baseline, which can only be trained from scratch.

U2 - 10.18653/v1/W19-5945

DO - 10.18653/v1/W19-5945

M3 - Conference contribution

SP - 392

EP - 398

BT - Proceedings of the 20th Annual SIGdial Meeting on Discourse and Dialogue

PB - Association for Computational Linguistics

T2 - 20th Annual SIGdial Meeting on Discourse and Dialogue 2019

Y2 - 11 September 2019 through 13 September 2019

ER -

User Evaluation of a Multi-dimensional Statistical Dialogue System

Abstract

Conference

Access to Document

Fingerprint

Cite this