Représentation et analyse automatique des discontinuités syntaxiques dans les corpus arborés en constituants du francais

Translated title of the contribution: Representation and parsing of syntactic discontinuities in French constituent treebanks

Maximin Coavoux, Benoît Crabbé

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The article introduces novel instanciations of three French constituent treebanks in which certain syntactic phenomena responsible for long-distance dependencies are represented with discontinuous constituents. Resulting trees are mildly context-sentitive structures, and can be modeled with, e.g. LCFRS. We show that such structures can be parsed efficiently, by introducing a neural transition-based discontinuous parser, that also performs morphological analysis and functional tagging. Our experiments show that the sparsity of these phenomena in French treebanks makes learning and evaluation of discontinuous structures difficult.
Translated title of the contributionRepresentation and parsing of syntactic discontinuities in French constituent treebanks
Original languageFrench
Title of host publicationActes de la 24e conférence sur le Traitement Automatique des Langues Naturelles
Place of PublicationOrléans, France
PublisherAssociation pour le Traitement Automatique des Langues (ATALA)
Pages77-92
Number of pages16
Publication statusPublished - 1 Jun 2017
EventAutomatic Processing of Natural Languages 2017 - Orleans, France
Duration: 26 Jun 201730 Jun 2017
http://taln2017.cnrs.fr/

Conference

ConferenceAutomatic Processing of Natural Languages 2017
Abbreviated titleTALN 2017
Country/TerritoryFrance
CityOrleans
Period26/06/1730/06/17
Internet address

Keywords / Materials (for Non-textual outputs)

  • Discontinuous constituents
  • parsing
  • deep learning
  • Constituants discontinus
  • analyse syntaxique

Fingerprint

Dive into the research topics of 'Representation and parsing of syntactic discontinuities in French constituent treebanks'. Together they form a unique fingerprint.

Cite this