Coarse-to-fine n-best parsing and MaxEnt discriminative reranking

Eugene Charniak*, Mark Johnson

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference proceeding contributionpeer-review

788 Citations (Scopus)

Abstract

Discriminative reranking is one method for constructing high-performance statistical parsers (Collins, 2000). A discriminative reranker requires a source of candidate parses for each sentence. This paper describes a simple yet novel method for constructing sets of 50-best parses based on a coarse-to-fine generative parser (Charniak, 2000). This method generates 50-best lists that are of substantially higher quality than previously obtainable. We used these parses as the input to a MaxEnt reranker (Johnson et al., 1999; Riezler et al., 2002) that selects the best parse from the set of parses for each sentence, obtaining an f-score of 91.0% on sentences of length 100 or less.

Original languageEnglish
Title of host publicationProceedings of the 43rd Annual Meeting of the Association for Computational Linguistics (ACL'2005)
Place of PublicationStroudsburg, PA
PublisherAssociation for Computational Linguistics (ACL)
Pages173-180
Number of pages8
ISBN (Print)1932432515, 9781932432510
Publication statusPublished - 2005
Externally publishedYes
Event43rd Annual Meeting of the Association for Computational Linguistics, ACL-05 - Ann Arbor, MI, United States
Duration: 25 Jun 200530 Jun 2005

Other

Other43rd Annual Meeting of the Association for Computational Linguistics, ACL-05
Country/TerritoryUnited States
CityAnn Arbor, MI
Period25/06/0530/06/05

Fingerprint

Dive into the research topics of 'Coarse-to-fine n-best parsing and MaxEnt discriminative reranking'. Together they form a unique fingerprint.

Cite this