Uncovering Machine Translationese: an experiment on 4 MT systems for English-French translations - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2020

Uncovering Machine Translationese: an experiment on 4 MT systems for English-French translations

Orphée de Clercq
  • Fonction : Auteur
Rudy Loock
Gert de Sutter
  • Fonction : Auteur
Bert Cappelle
Koen Plevoets
  • Fonction : Auteur

Résumé

The aim of this presentation is to discuss the linguistic features of machine-translated texts in comparison with original texts in order to uncover what has been called “machine translationese” (e.g. Daems et al. 2017). Using a corpus-based statistical approach, namely, the Principal Component Analysis technique, 4 MT systems have been investigated for English to French translations of press texts: 1 Statistical MT (SMT) and 3 Neural MT (NMT) systems, namely DeepL, Google Translate, and the European Commission’s eTranslation MT tool, in both its SMT and NMT versions. In particular, to complement a previous study on language-specific features (e.g. derived adverbs, existential constructions, coordinator et, preposition avec, see Loock 2018), a series of language-independent linguistic features were extracted for each text, ranging from superficial text characteristics such as the average word and sentence length, to frequencies of closed-class lexical categories and measures of lexical diversity.The final aim is to uncover linguistic features in MT texts that clearly deviate from the expected norms in original French.
Fichier non déposé

Dates et versions

hal-02454668 , version 1 (24-01-2020)

Identifiants

  • HAL Id : hal-02454668 , version 1

Citer

Orphée de Clercq, Rudy Loock, Gert de Sutter, Bert Cappelle, Koen Plevoets. Uncovering Machine Translationese: an experiment on 4 MT systems for English-French translations. Journée d'études #TQ2020 "Traduction & Qualité : biotraduction et traduction automatique", Université de Lille & UMR STL du CNRS, Jan 2020, Roubaix, France. ⟨hal-02454668⟩
208 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More