en
Master
Open access
English

Assessing the quality and usability for gisting purposes of a French into Spanish neural machine translation raw output for a legal text

Master program titleMaîtrise universitaire en traduction : mention Technologies de la traduction
Defense date2024
Abstract

This research had a dual purpose: 1) establishing whether the quality (adequacy and fluidity) of a French into Spanish raw machine translation (MT) output for a legal text was satisfactory from a professional perspective for gisting purposes for end users; and 2) determining whether that MT output was usable for gisting purposes from the end users' perspective. The results suggest that the raw MT output had a satisfactory quality for gisting purposes for end users from a professional perspective. In addition, the raw MT output would be usable from the end users’ perspective as participants performed the proposed reading comprehension task (a gap-filling task) efficiently and effectively and were satisfied with the raw MT output. Regardless these results, there were adequacy and fluidity issues in the raw MT output, some of which seemed to have had a negative impact on reading comprehension at both the segment and whole text levels. Hence, there is still room for improvement, especially considering we dealt with a legal text.

engfre
Keywords
  • Adequacy
  • Effectiveness
  • Efficiency
  • Fluidity
  • Gap-filling
  • Gisting
  • Legal text
  • Machine translation
  • Quality
  • Satisfactory
  • Usability
  • Adéquation
  • Assimilation
  • Efficacité
  • Effectivité
  • Exercice à trous
  • Fluidité
  • Qualité
  • Satisfaisant
  • Traduction automatique
  • Texte juridique
  • Utilisabilité
Citation (ISO format)
LOPEZ, Sonia Viviana. Assessing the quality and usability for gisting purposes of a French into Spanish neural machine translation raw output for a legal text. 2024.
Main files (1)
Master thesis
accessLevelPublic
Identifiers
  • PID : unige:175566
42views
27downloads

Technical informations

Creation03/12/2024 1:31:01 PM
First validation03/13/2024 7:19:29 AM
Update time03/28/2024 7:23:24 AM
Status update03/28/2024 7:23:24 AM
Last indexation05/06/2024 6:09:08 PM
All rights reserved by Archive ouverte UNIGE and the University of GenevaunigeBlack