Transforming Projective Bilexical Dependency Grammars into efficiently-parsable CFGs with Unfold-Fold

Mark Johnson*

*Corresponding author for this work

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

26 Citations (Scopus)

Abstract

This paper shows how to use the Unfold-Fold transformation to transform Projective Bilexical Dependency Grammars (PBDGs) into ambiguity-preserving weakly equivalent Context-Free Grammars (CFGs). These CFGs can be parsed in O(n 3) time using a CKY algorithm with appropriate indexing, rather than the O(n 5) time required by a naive encoding. Informally, using the CKY algorithm with such a CFG mimics the steps of the Eisner-Satta O(n 3) PBDG parsing algorithm. This transformation makes all of the techniques developed for CFGs available to PBDGs. We demonstrate this by describing a maximum posterior parse decoder for PBDGs.

Original languageEnglish
Title of host publicationACL 2007 - Proceedings of the 45th Annual Meeting of the Association for Computational Linguistics
Place of PublicationStroudsburg, PA
PublisherAssociation for Computational Linguistics (ACL)
Pages168-175
Number of pages8
ISBN (Print)9781932432862
Publication statusPublished - 2007
Externally publishedYes
Event45th Annual Meeting of the Association for Computational Linguistics, ACL 2007 - Prague, Czech Republic
Duration: 23 Jun 200730 Jun 2007

Other

Other45th Annual Meeting of the Association for Computational Linguistics, ACL 2007
Country/TerritoryCzech Republic
CityPrague
Period23/06/0730/06/07

Fingerprint

Dive into the research topics of 'Transforming Projective Bilexical Dependency Grammars into efficiently-parsable CFGs with Unfold-Fold'. Together they form a unique fingerprint.

Cite this