Use this url to cite publication: https://hdl.handle.net/20.500.12259/56635
Lithuanian dependency parsing with rich morphological features
Type of publication
Straipsnis konferencijos medžiagoje Scopus duomenų bazėje / Article in conference proceedings in Scopus database (P1a2)
Author(s)
Author | Affiliation | |
---|---|---|
Kauno technologijos universitetas | LT | |
Nivre, Joakim | ||
Title [en]
Lithuanian dependency parsing with rich morphological features
Is part of
SPMRL'2013 : proceedings of the fourth workshop on statistical parsing of morphologically rich languages, October 18-21, 2013, Seattle, Washington, USA. Stroudsburg : Association for computational linguistics, 2013
Date Issued
Date |
---|
2013 |
Publisher
Stroudsburg : Association for computational linguistics, 2013
Is Referenced by
Extent
p. 12-21
Abstract (en)
We present the first statistical dependency parsing results for Lithuanian, a morphologically rich language in the Baltic branch of the Indo-European family. Using a greedy transition-based parser, we obtain a labeled attachment score of 74.7 with gold morphology and 68.1 with predicted morphology (77.8 and 72.8 unlabeled). We investigate the usefulness of different features and find that rich morphological features improve parsing accuracy significantly, by 7.5 percentage points with gold features and 5.6 points with predicted features. As expected, CASE is the single most important morphological feature, but virtually all available features bring some improvement, especially under the gold condition.
Type of document
type::text::journal::journal article::research article
Language
Anglų / English (en)
Coverage Spatial
Jungtinės Amerikos Valstijos / United States of America (US)
ISBN (of the container)
9781937284978
Other Identifier(s)
VDU02-000014467