Surface Realisation Using Full Delexicalisation - Archive ouverte HAL Access content directly
Conference Papers Year : 2019

Surface Realisation Using Full Delexicalisation

Abstract

Surface realisation (SR) maps a meaning representation to a sentence and can be viewed as consisting of three subtasks: word ordering, morphological inflection and contraction generation (e.g., clitic attachment in Portuguese or elision in French). We propose a modular approach to surface realisation which models each of these components separately, and evaluate our approach on the 10 languages covered by the SR'18 Surface Realisation Shared Task shallow track. We provide a detailed evaluation of how word order, morphological realisa-tion and contractions are handled by the model and an analysis of the differences in word ordering performance across languages.
Fichier principal
Vignette du fichier
2019-EMNLP-Anastasia.pdf (209.89 Ko) Télécharger le fichier
Origin Files produced by the author(s)
Loading...

Dates and versions

hal-02277069 , version 1 (03-09-2019)

Identifiers

Cite

Anastasia Shimorina, Claire Gardent. Surface Realisation Using Full Delexicalisation. 2019 Conference on Empirical Methods in Natural Language Processing and 9th International Joint Conference on Natural Language Processing, Nov 2019, Hong Kong, China. pp.3084-3094, ⟨10.18653/v1/D19-1305⟩. ⟨hal-02277069⟩
86 View
323 Download

Altmetric

Share

Gmail Mastodon Facebook X LinkedIn More