Quick Search:

uzh logo
Browse by:
bullet
bullet
bullet
bullet

Zurich Open Repository and ArchiveĀ 

Permanent URL to this publication: http://dx.doi.org/10.5167/uzh-63507

Schneider, G (2012). Using semantic resources to improve a syntactic dependency parser. In: LREC 2012 Conference Workshop "Semantic Relations II", Istanbul, Turkey, 22 May 2012 - 22 May 2012, 67-76.

[img]
Preview
Published Version
PDF
209kB

Abstract

Probabilistic syntactic parsing has made rapid progress, but is reaching a performance ceiling. More semantic resources need to be included. We exploit a number of semantic resources to improve parsing accuracy of a dependency parser. We compare semantic lexica on this task, then we extend the back-off chain by punishing underspecified decisions. Further, a simple distributional semantics approach is tested. Selectional restrictions are employed to boost interpretations that are semantically plausible. We also show that self-training can improve parsing even without needing a re-ranker, as we can rely on a sufficiently good estimation of parsing accuracy. Parsing large amounts of data and using it in self-training allows us to learn world knowledge from the distribution of syntactic relation. We show that the performance of the parser considerably improves due to our extensions.

Item Type:Conference or Workshop Item (Paper), refereed, original work
Communities & Collections:06 Faculty of Arts > English Department
06 Faculty of Arts > Institute of Computational Linguistics
DDC:820 English & Old English literatures
410 Linguistics
000 Computer science, knowledge & systems
Uncontrolled Keywords:Exploitation of semantic resources for NLP applications Syntactic parsing WordNet and WordNet-like resources Self-training Distributional semantics
Language:English
Event End Date:22 May 2012
Deposited On:19 Jul 2012 06:54
Last Modified:02 Nov 2012 00:58
Official URL:http://www.lrec-conf.org/proceedings/lrec2012/workshops/10.Semantic%20Relations%20II%20Proceedings.pdf
Citations:Google Scholarā„¢

Users (please log in): suggest update or correction for this item

Repository Staff Only: item control page