Enforcing Subcategorization Constraints in a Parser Using Sub-parses Recombining

Abstract : Treebanks are not large enough to adequately model subcategorization frames of predicative lexemes, which is an important source of lexico-syntactic constraints for parsing. As a consequence, parsers trained on such treebanks usually make mistakes when selecting the arguments of predicative lexemes. In this paper, we propose an original way to correct subcategorization errors by combining sub-parses of a sentence S that appear in the list of the n-best parses of S. The subcategorization information comes from three different resources, the first one is extracted from a treebank, the second one is computed on a large corpora and the third one is an existing syntactic lexicon. Experiments on the French Treebank showed a 15.24% reduction of erroneous subcategorization frames (SF) selections for verbs as well as a relative decrease of the error rate of 4% Labeled Accuracy Score on the state of the art parser on this treebank.
Document type :
Conference papers
Complete list of metadatas

Cited literature [26 references]  Display  Hide  Download

https://hal.inria.fr/hal-00936492
Contributor : Benoît Sagot <>
Submitted on : Sunday, January 26, 2014 - 3:36:12 PM
Last modification on : Friday, January 4, 2019 - 5:33:24 PM
Long-term archiving on : Saturday, April 26, 2014 - 10:15:52 PM

File

N13-1024.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00936492, version 1

Collections

Citation

Seyed Abolghasem Mirroshandel, Alexis Nasr, Benoît Sagot. Enforcing Subcategorization Constraints in a Parser Using Sub-parses Recombining. NAACL 2013 - Conference of the North American Chapter of the Association for Computational Linguistics, Jun 2013, Atlanta, United States. ⟨hal-00936492⟩

Share

Metrics

Record views

687

Files downloads

176