A minimum message length approach for argument interpretation

Ingrid Zukerman; Sarah George

A minimum message length approach for argument interpretation

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

Abstract

We describe a mechanism which receives as input a segmented argument composed of NL sentences, and generates an interpretation. Our mechanism relies on the Minimum Message Length Principle for the selection of an interpretation among candidate options. This enables our mechanism to cope with noisy input in terms of wording, beliefs and argument structure; and reduces its reliance on a particular knowledge representation. The performance of our system was evaluated by distorting automatically generated arguments, and passing them to the system for interpretation. In 75% of the cases, the interpretations produced by the system matched precisely or almost-precisely the representation of the original arguments.

Original language	English
Title of host publication	Third SIGdial Workshop on Discourse and Dialogue: Proceedings of the Workshop
Editors	K Jokinen, S McRoy
Place of Publication	New Brunswick USA
Publisher	Association for Computational Linguistics (ACL)
Pages	211-220
Number of pages	10
Publication status	Published - 2002
Event	SIGdial Workshop on Discourse and Dialog 2002 - Philadelphia, United States of America Duration: 11 Jul 2002 → 12 Jul 2002 Conference number: 3rd

Conference

Conference	SIGdial Workshop on Discourse and Dialog 2002
Abbreviated title	SIGDIAL 2002
Country/Territory	United States of America
City	Philadelphia
Period	11/07/02 → 12/07/02

Cite this

@inproceedings{2a66788bdb2b4d719d9d24a93c473595,

title = "A minimum message length approach for argument interpretation",

abstract = "We describe a mechanism which receives as input a segmented argument composed of NL sentences, and generates an interpretation. Our mechanism relies on the Minimum Message Length Principle for the selection of an interpretation among candidate options. This enables our mechanism to cope with noisy input in terms of wording, beliefs and argument structure; and reduces its reliance on a particular knowledge representation. The performance of our system was evaluated by distorting automatically generated arguments, and passing them to the system for interpretation. In 75% of the cases, the interpretations produced by the system matched precisely or almost-precisely the representation of the original arguments.",

author = "Ingrid Zukerman and Sarah George",

note = "Funding Information: This research was supported in part by Australian Research Council grant A49927212. Publisher Copyright: {\textcopyright} SIGDIAL 2002 .All right reserved.; SIGdial Workshop on Discourse and Dialog 2002, SIGDIAL 2002 ; Conference date: 11-07-2002 Through 12-07-2002",

year = "2002",

language = "English",

pages = "211--220",

editor = "K Jokinen and S McRoy",

booktitle = "Third SIGdial Workshop on Discourse and Dialogue: Proceedings of the Workshop",

publisher = "Association for Computational Linguistics (ACL)",

}

Zukerman, I & George, S 2002, A minimum message length approach for argument interpretation. in K Jokinen & S McRoy (eds), Third SIGdial Workshop on Discourse and Dialogue: Proceedings of the Workshop. Association for Computational Linguistics (ACL), New Brunswick USA, pp. 211-220, SIGdial Workshop on Discourse and Dialog 2002, Philadelphia, Pennsylvania, United States of America, 11/07/02.

A minimum message length approach for argument interpretation. / Zukerman, Ingrid; George, Sarah.

Third SIGdial Workshop on Discourse and Dialogue: Proceedings of the Workshop. ed. / K Jokinen; S McRoy. New Brunswick USA : Association for Computational Linguistics (ACL), 2002. p. 211-220.

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

TY - GEN

T1 - A minimum message length approach for argument interpretation

AU - Zukerman, Ingrid

AU - George, Sarah

N1 - Conference code: 3rd

PY - 2002

Y1 - 2002

N2 - We describe a mechanism which receives as input a segmented argument composed of NL sentences, and generates an interpretation. Our mechanism relies on the Minimum Message Length Principle for the selection of an interpretation among candidate options. This enables our mechanism to cope with noisy input in terms of wording, beliefs and argument structure; and reduces its reliance on a particular knowledge representation. The performance of our system was evaluated by distorting automatically generated arguments, and passing them to the system for interpretation. In 75% of the cases, the interpretations produced by the system matched precisely or almost-precisely the representation of the original arguments.

AB - We describe a mechanism which receives as input a segmented argument composed of NL sentences, and generates an interpretation. Our mechanism relies on the Minimum Message Length Principle for the selection of an interpretation among candidate options. This enables our mechanism to cope with noisy input in terms of wording, beliefs and argument structure; and reduces its reliance on a particular knowledge representation. The performance of our system was evaluated by distorting automatically generated arguments, and passing them to the system for interpretation. In 75% of the cases, the interpretations produced by the system matched precisely or almost-precisely the representation of the original arguments.

UR - http://www.scopus.com/inward/record.url?scp=60049085439&partnerID=8YFLogxK

M3 - Conference Paper

SP - 211

EP - 220

BT - Third SIGdial Workshop on Discourse and Dialogue: Proceedings of the Workshop

A2 - Jokinen, K

A2 - McRoy, S

PB - Association for Computational Linguistics (ACL)

CY - New Brunswick USA

T2 - SIGdial Workshop on Discourse and Dialog 2002

Y2 - 11 July 2002 through 12 July 2002

ER -

A minimum message length approach for argument interpretation

Abstract

Conference

Other files and links

Cite this