A Comparison between NMT and PBSMT Performance for Translating Noisy User-Generated Content - Laboratoire d'Informatique pour la Mécanique et les Sciences de l'Ingénieur Accéder directement au contenu
Communication Dans Un Congrès Année : 2019

A Comparison between NMT and PBSMT Performance for Translating Noisy User-Generated Content

Résumé

This work compares the performances achieved by Phrase-Based Statistical Ma- chine Translation systems (PBSMT) and attention-based Neural Machine Transla- tion systems (NMT) when translating User Generated Content (UGC), as encountered in social medias, from French to English. We show that, contrary to what could be ex- pected, PBSMT outperforms NMT when translating non-canonical inputs. Our error analysis uncovers the specificities of UGC that are problematic for sequential NMT architectures and suggests new avenue for improving NMT models.
Fichier principal
Vignette du fichier
nodalida2019.pdf (1.39 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02270524 , version 1 (25-08-2019)

Identifiants

  • HAL Id : hal-02270524 , version 1

Citer

José Carlos Rosales Nunez, Djamé Seddah, Guillaume Wisniewski. A Comparison between NMT and PBSMT Performance for Translating Noisy User-Generated Content. The 22nd Nordic Conference on Computational Linguistics (NoDaLiDa’19), Sep 2019, Turku, Finland. ⟨hal-02270524⟩
283 Consultations
192 Téléchargements

Partager

Gmail Facebook X LinkedIn More