Spanning Tree Methods for Discriminative Training of Dependency Parsers

Simple item page
dc.contributor.authorMcDonald, Ryan
dc.contributor.authorPereira, Fernando C.N.
dc.contributor.authorPereira, Fernando C.N.
dc.date2023-05-17T00:04:51.000
dc.date.accessioned2023-05-22T12:57:36Z
dc.date.available2023-05-22T12:57:36Z
dc.date.issued2006-01-01
dc.date.submitted2006-10-26T07:42:30-07:00
dc.description.abstractUntyped dependency parsing can be viewed as the problem of finding maximum spanning trees (MSTs) in directed graphs. Using this representation, the Eisner (1996) parsing algorithm is sufficient for searching the space of projective trees. More importantly, the representation is extended naturally to non-projective parsing using Chu-Liu-Edmonds (Chu and Liu, 1965; Edmonds, 1967) MST algorithm. These efficient parse search methods support large-margin discriminative training methods for learning dependency parsers. We evaluate these methods experimentally on the English and Czech treebanks.
dc.description.commentsUniversity of Pennsylvania Department of Computer and Information Science Technical Report No. MS-CIS-06-11.
dc.identifier.urihttps://repository.upenn.edu/handle/20.500.14332/7490
dc.legacy.articleid1056
dc.legacy.fulltexturlhttps://repository.upenn.edu/cgi/viewcontent.cgi?article=1056&context=cis_reports&unstamped=1
dc.source.issue55
dc.source.journalTechnical Reports (CIS)
dc.source.statuspublished
dc.titleSpanning Tree Methods for Discriminative Training of Dependency Parsers
dc.typeReport
digcom.contributor.authorMcDonald, Ryan
digcom.contributor.authorCrammer, Koby
digcom.contributor.authorisAuthorOfPublication|email:pereira@cis.upenn.edu|institution:University of Pennsylvania|Pereira, Fernando C.N.
digcom.identifiercis_reports/55
digcom.identifier.contextkey219216
digcom.identifier.submissionpathcis_reports/55
digcom.typereport
dspace.entity.typePublication
relation.isAuthorOfPublication1e403b95-e31d-41cd-a1fe-72bde154dec9
relation.isAuthorOfPublication1e403b95-e31d-41cd-a1fe-72bde154dec9
relation.isAuthorOfPublication.latestForDiscovery1e403b95-e31d-41cd-a1fe-72bde154dec9
upenn.schoolDepartmentCenterTechnical Reports (CIS)
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
MS_CIS_05_11.pdf
Size:
177.55 KB
Format:
Adobe Portable Document Format
Download
Collection
Reports