Efficient probabilistic top-down and left-corner parsing

Brian Roark; Mark Johnson

doi:10.3115/1034678.1034743

Efficient probabilistic top-down and left-corner parsing

Brian Roark, Mark Johnson

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

24 Downloads (Pure)

Abstract

This paper examines efficient predictive broad, coverage parsing without dynamic programming. In contrast to bottom-up methods, depth-first top-down parsing produces partial parses that are fully connected trees spanning the entire left context, from which any kind of non-local dependency or partial semantic interpretation can in principle be read. We contrast two predictive parsing approaches, top-down and left-corner parsing, and find both to be viable. In addition, we find that enhancement with non-local information not only improves parser accuracy, but also substantially improves the search efficiency.

Original language	English
Title of host publication	Proceedings of the 37th Annual Meeting of the Association for Computational Linguistics
Place of Publication	San Francisco
Publisher	Association for Computational Linguistics
Pages	421-428
Number of pages	8
ISBN (Print)	1558606092
DOIs	https://doi.org/10.3115/1034678.1034743
Publication status	Published - 1999
Externally published	Yes
Event	Annual Meeting of the Association for Computational Linguistics (37th : 1999) - University of Maryland, College Park, United States Duration: 20 Jun 1999 → 26 Jun 1999

Conference

Conference	Annual Meeting of the Association for Computational Linguistics (37th : 1999)
Country/Territory	United States
City	College Park
Period	20/06/99 → 26/06/99

Bibliographical note

Copyright the Publisher 1999. Version archived for private and non-commercial use with the permission of the author/s and according to publisher conditions. For further rights please contact the publisher.

Access to Document

10.3115/1034678.1034743

Publisher version (open access)Final published version, 648 KBLicence: CC BY-NC-SA

Cite this

@inproceedings{f0b94327fee1477fb19554791043566e,

title = "Efficient probabilistic top-down and left-corner parsing",

abstract = "This paper examines efficient predictive broad, coverage parsing without dynamic programming. In contrast to bottom-up methods, depth-first top-down parsing produces partial parses that are fully connected trees spanning the entire left context, from which any kind of non-local dependency or partial semantic interpretation can in principle be read. We contrast two predictive parsing approaches, top-down and left-corner parsing, and find both to be viable. In addition, we find that enhancement with non-local information not only improves parser accuracy, but also substantially improves the search efficiency.",

author = "Brian Roark and Mark Johnson",

note = "Copyright the Publisher 1999. Version archived for private and non-commercial use with the permission of the author/s and according to publisher conditions. For further rights please contact the publisher. ; Annual Meeting of the Association for Computational Linguistics (37th : 1999) ; Conference date: 20-06-1999 Through 26-06-1999",

year = "1999",

doi = "10.3115/1034678.1034743",

language = "English",

isbn = "1558606092 ",

pages = "421--428",

booktitle = "Proceedings of the 37th Annual Meeting of the Association for Computational Linguistics",

publisher = "Association for Computational Linguistics",

}

Roark, B & Johnson, M 1999, Efficient probabilistic top-down and left-corner parsing. in Proceedings of the 37th Annual Meeting of the Association for Computational Linguistics. Association for Computational Linguistics, San Francisco, pp. 421-428, Annual Meeting of the Association for Computational Linguistics (37th : 1999), College Park, Maryland, United States, 20/06/99. https://doi.org/10.3115/1034678.1034743

TY - GEN

T1 - Efficient probabilistic top-down and left-corner parsing

AU - Roark, Brian

AU - Johnson, Mark

N1 - Copyright the Publisher 1999. Version archived for private and non-commercial use with the permission of the author/s and according to publisher conditions. For further rights please contact the publisher.

PY - 1999

Y1 - 1999

N2 - This paper examines efficient predictive broad, coverage parsing without dynamic programming. In contrast to bottom-up methods, depth-first top-down parsing produces partial parses that are fully connected trees spanning the entire left context, from which any kind of non-local dependency or partial semantic interpretation can in principle be read. We contrast two predictive parsing approaches, top-down and left-corner parsing, and find both to be viable. In addition, we find that enhancement with non-local information not only improves parser accuracy, but also substantially improves the search efficiency.

AB - This paper examines efficient predictive broad, coverage parsing without dynamic programming. In contrast to bottom-up methods, depth-first top-down parsing produces partial parses that are fully connected trees spanning the entire left context, from which any kind of non-local dependency or partial semantic interpretation can in principle be read. We contrast two predictive parsing approaches, top-down and left-corner parsing, and find both to be viable. In addition, we find that enhancement with non-local information not only improves parser accuracy, but also substantially improves the search efficiency.

U2 - 10.3115/1034678.1034743

DO - 10.3115/1034678.1034743

M3 - Conference proceeding contribution

SN - 1558606092

SP - 421

EP - 428

BT - Proceedings of the 37th Annual Meeting of the Association for Computational Linguistics

PB - Association for Computational Linguistics

CY - San Francisco

T2 - Annual Meeting of the Association for Computational Linguistics (37th : 1999)

Y2 - 20 June 1999 through 26 June 1999

ER -

Efficient probabilistic top-down and left-corner parsing

Abstract

Conference

Bibliographical note

Access to Document

Fingerprint

Cite this