Idiomatikotasunaren karakterizazio automatikoaizena+aditza

  1. Antton Gurrutxaga Hernaiz
  2. Iñaki Alegria Loinaz
  3. Xabier Artola Zubillaga
Journal:
Ekaia: Euskal Herriko Unibertsitateko zientzi eta teknologi aldizkaria

ISSN: 0214-9001

Year of publication: 2016

Issue Title: 2013-2014 Euskal tesien 10 pasarte

Issue: 1

Pages: 47-68

Type: Article

DOI: 10.1387/EKAIA.14544 DIALNET GOOGLE SCHOLAR lock_openOpen access editor

More publications in: Ekaia: Euskal Herriko Unibertsitateko zientzi eta teknologi aldizkaria

Abstract

The goal of this research is to develop and experimentally test different techniques for the automatic extraction of phraseological units (PUs) of noun+verb structure in Basque and for their characterization according to the idiomaticity level. Idiomaticity is considered the defining feature of the concept of phraseological unit (PU), ande we have measured its following components: institutionalization (statistical idiosyncrasy), semantic non-compositionality, morphosyntact ic fixedness and lexical fixedness. The results show that the standard cooccurence techniques are significantly ourtperformed by semantic measures, and, to a lower extent, by measures of morphosyntactic flexibility. The results of lexical flexibility are poorer than expected. Finally, we obtain experimental evidence for several predictions of phraseological theory.