Vector Spaces for Quantifying Disparity of Multiword Expressions in Annotated Text - Laboratoire Interdisciplinaire des Sciences du Numérique
Conference Papers Year : 2024

Vector Spaces for Quantifying Disparity of Multiword Expressions in Annotated Text

Espaces Vectoriels pour la Quantification de la Disparité d'Expressions Polylexicales dans les Textes Annotés

Abstract

Multiword Expressions (MWEs) make a good case study for linguistic diversity due to their idiosyncratic nature. Defining MWE canonical forms as types, diversity may be measured notably through disparity, based on pairwise distances between types. To this aim, we train static MWE-aware word embeddings for verbal MWEs in 14 languages, and we show interesting properties of these vector spaces. We use these vector spaces to implement the so-called functional diversity measure. We apply this measure to the results of several MWE identification systems. We find that, although MWE vector spaces are meaningful at a local scale, the disparity measure aggregating them at a global scale strongly correlates with the number of types, which questions its usefulness in presence of simpler diversity metrics such as variety. We make the vector spaces we generated available.
Fichier principal
Vignette du fichier
ESTEVE_SAVARY_LAVERGNE_ACL-SRW.pdf (3.02 Mo) Télécharger le fichier
Origin Files produced by the author(s)
licence

Dates and versions

hal-04660179 , version 1 (23-07-2024)

Licence

Identifiers

  • HAL Id : hal-04660179 , version 1

Cite

Louis Estève, Agata Savary, Thomas Lavergne. Vector Spaces for Quantifying Disparity of Multiword Expressions in Annotated Text. Association for Computational Linguistics - Student Research Workshop, Aug 2024, Bangkok, Thailand. ⟨hal-04660179⟩
59 View
31 Download

Share

More