Generating Referring Expressions in Open Domains

Advaith Siddharthan; Ann Copestake

Generating Referring Expressions in Open Domains

Advaith Siddharthan, Ann Copestake

Computing Science

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

Abstract

We present an algorithm for generating referring expressions in open domains. Existing algorithms work at the semantic level and assume the availability of a classication for attributes, which is only feasible for restricted domains. Our alternative works at the realisation level, relies on Word-Net synonym and antonym sets, and gives equivalent results on the examples cited in the literature and improved results for examples that prior approaches cannot handle. We believe that ours is also the first algorithm that allows for the incremental incorporation of relations. We present a novel corpus-evaluation using referring expressions from
the Penn Wall Street Journal Treebank.

Original language	English
Title of host publication	Proceedings of the 42nd Annual Meeting on Association for Computational Linguistics (ACL)
Place of Publication	Barcelona, Spain
Publication status	Published - 2004
Event	42nd Annual Meeting on Association for Computational Linguistics - Barcelona, Spain Duration: 21 Jul 2004 → 26 Jul 2004

Conference

Conference	42nd Annual Meeting on Association for Computational Linguistics
Country/Territory	Spain
City	Barcelona
Period	21/07/04 → 26/07/04

Bibliographical note

Proceedings of the 42nd Annual Meeting on Association for Computational Linguistics, 21-27 July, 2004, Barcelona, Spain

Access to Document

http://aclweb.org/anthology/P/P04/P04-1052.pdf

Cite this

@inproceedings{ed196dbabaa54a2c8b549ace33674619,

title = "Generating Referring Expressions in Open Domains",

abstract = "We present an algorithm for generating referring expressions in open domains. Existing algorithms work at the semantic level and assume the availability of a classication for attributes, which is only feasible for restricted domains. Our alternative works at the realisation level, relies on Word-Net synonym and antonym sets, and gives equivalent results on the examples cited in the literature and improved results for examples that prior approaches cannot handle. We believe that ours is also the first algorithm that allows for the incremental incorporation of relations. We present a novel corpus-evaluation using referring expressions from the Penn Wall Street Journal Treebank. ",

author = "Advaith Siddharthan and Ann Copestake",

note = "Proceedings of the 42nd Annual Meeting on Association for Computational Linguistics, 21-27 July, 2004, Barcelona, Spain ; 42nd Annual Meeting on Association for Computational Linguistics ; Conference date: 21-07-2004 Through 26-07-2004",

year = "2004",

language = "English",

booktitle = "Proceedings of the 42nd Annual Meeting on Association for Computational Linguistics (ACL)",

}

TY - GEN

T1 - Generating Referring Expressions in Open Domains

AU - Siddharthan, Advaith

AU - Copestake, Ann

N1 - Proceedings of the 42nd Annual Meeting on Association for Computational Linguistics, 21-27 July, 2004, Barcelona, Spain

PY - 2004

Y1 - 2004

N2 - We present an algorithm for generating referring expressions in open domains. Existing algorithms work at the semantic level and assume the availability of a classication for attributes, which is only feasible for restricted domains. Our alternative works at the realisation level, relies on Word-Net synonym and antonym sets, and gives equivalent results on the examples cited in the literature and improved results for examples that prior approaches cannot handle. We believe that ours is also the first algorithm that allows for the incremental incorporation of relations. We present a novel corpus-evaluation using referring expressions from the Penn Wall Street Journal Treebank.

AB - We present an algorithm for generating referring expressions in open domains. Existing algorithms work at the semantic level and assume the availability of a classication for attributes, which is only feasible for restricted domains. Our alternative works at the realisation level, relies on Word-Net synonym and antonym sets, and gives equivalent results on the examples cited in the literature and improved results for examples that prior approaches cannot handle. We believe that ours is also the first algorithm that allows for the incremental incorporation of relations. We present a novel corpus-evaluation using referring expressions from the Penn Wall Street Journal Treebank.

M3 - Published conference contribution

BT - Proceedings of the 42nd Annual Meeting on Association for Computational Linguistics (ACL)

CY - Barcelona, Spain

T2 - 42nd Annual Meeting on Association for Computational Linguistics

Y2 - 21 July 2004 through 26 July 2004

ER -

Generating Referring Expressions in Open Domains

Abstract

Conference

Bibliographical note

Access to Document

Fingerprint

Cite this