Building a morphological and syntactic lexicon by merging various linguistic resources

Miguel Molinero 1 Benoît Sagot 2 Lionel Nicolas 3
2 ALPAGE - Analyse Linguistique Profonde à Grande Echelle ; Large-scale deep linguistic processing
Inria Paris-Rocquencourt, UPD7 - Université Paris Diderot - Paris 7
3 Laboratoire d'Informatique, Signaux, et Systèmes de Sophia-Antipolis (I3S) / Equipe RL
Laboratoire I3S - MDSC - Modèles Discrets pour les Systèmes Complexes
Abstract : This paper shows how large-coverage morphological and syntactic NLP lexicons can be developed by interpreting, converting to a common format and merging existing lexical resources. Applied on Spanish, this allowed us to build a morphological and syntactic lexicon, the Leffe. It relies on the Alexina framework, originally developed together 200with the French lexicon Lefff. We describe how the input resources --two morphological and two syntactic lexicons -- were converted into Alexina lexicons and merged. A preliminary evaluation shows that merging different sources of lexical information is indeed a good approach to improve the development speed, the coverage and the precision of linguistic resources.
Document type :
Conference papers
Liste complète des métadonnées

https://hal.inria.fr/hal-00793048
Contributor : Brigitte Briot <>
Submitted on : Thursday, February 21, 2013 - 2:40:41 PM
Last modification on : Friday, January 4, 2019 - 5:33:24 PM

Identifiers

  • HAL Id : hal-00793048, version 1

Citation

Miguel Molinero, Benoît Sagot, Lionel Nicolas. Building a morphological and syntactic lexicon by merging various linguistic resources. NODALIDA 2009 - the 17th Nordic Conference of Computational Linguistics, May 2009, Odense, Denmark. ⟨hal-00793048⟩

Share

Metrics

Record views

199