Building a robust dialogue system with limited data

Sharon J. Goldwater; Elizabeth Owen Bratt; Jean Mark Gawron; John Dowding

Building a robust dialogue system with limited data

Sharon J. Goldwater, Elizabeth Owen Bratt, Jean Mark Gawron, John Dowding

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

Abstract

We describe robustness techniques used in the CommandTalk system at the recognition level, the parsing level, and the dialogue level, and how these were influenced by the lack of domain data. We used interviews with subject matter experts (SME's) to develop a single grammar for recognition, understanding, and generation, this eliminating the need for a robust parser. We broadened the coverage of the recognition grammar by allowing word insertions and deletions, and we implemented clarification and correction subdialogues to increase robustness at the dialogue level. We discuss the applicability of these techniques to other domains.

Original language	English
Title of host publication	Proceedings of the ANLP-NAACL 2000 Workshop on Conversational Systems
Place of Publication	Stroudsburg, PA, USA
Publisher	Association for Computational Linguistics
Pages	61-65
Number of pages	5
Publication status	Published - 2000

Access to Document

W00 0312Final published version, 347 KB

http://www.aclweb.org/anthology-new/W/W00/W00-0312.pdf

Cite this

@inproceedings{851897ce239140f38094eab62552b6cb,

title = "Building a robust dialogue system with limited data",

abstract = "We describe robustness techniques used in the CommandTalk system at the recognition level, the parsing level, and the dialogue level, and how these were influenced by the lack of domain data. We used interviews with subject matter experts (SME's) to develop a single grammar for recognition, understanding, and generation, this eliminating the need for a robust parser. We broadened the coverage of the recognition grammar by allowing word insertions and deletions, and we implemented clarification and correction subdialogues to increase robustness at the dialogue level. We discuss the applicability of these techniques to other domains.",

author = "Goldwater, {Sharon J.} and Bratt, {Elizabeth Owen} and Gawron, {Jean Mark} and John Dowding",

year = "2000",

language = "English",

pages = "61--65",

booktitle = "Proceedings of the ANLP-NAACL 2000 Workshop on Conversational Systems",

publisher = "Association for Computational Linguistics",

}

TY - GEN

T1 - Building a robust dialogue system with limited data

AU - Goldwater, Sharon J.

AU - Bratt, Elizabeth Owen

AU - Gawron, Jean Mark

AU - Dowding, John

PY - 2000

Y1 - 2000

N2 - We describe robustness techniques used in the CommandTalk system at the recognition level, the parsing level, and the dialogue level, and how these were influenced by the lack of domain data. We used interviews with subject matter experts (SME's) to develop a single grammar for recognition, understanding, and generation, this eliminating the need for a robust parser. We broadened the coverage of the recognition grammar by allowing word insertions and deletions, and we implemented clarification and correction subdialogues to increase robustness at the dialogue level. We discuss the applicability of these techniques to other domains.

AB - We describe robustness techniques used in the CommandTalk system at the recognition level, the parsing level, and the dialogue level, and how these were influenced by the lack of domain data. We used interviews with subject matter experts (SME's) to develop a single grammar for recognition, understanding, and generation, this eliminating the need for a robust parser. We broadened the coverage of the recognition grammar by allowing word insertions and deletions, and we implemented clarification and correction subdialogues to increase robustness at the dialogue level. We discuss the applicability of these techniques to other domains.

M3 - Conference contribution

SP - 61

EP - 65

BT - Proceedings of the ANLP-NAACL 2000 Workshop on Conversational Systems

PB - Association for Computational Linguistics

CY - Stroudsburg, PA, USA

ER -

Building a robust dialogue system with limited data

Abstract

Access to Document

Fingerprint

Cite this