Master
OA Policy
English

An evaluation of part-of-speech taggers for French

ContributorsMattiuzzi, Silvia
Master program titleMaîtrise universitaire en traduction et technologie
Defense date2021
Abstract

Annotated corpora are widely employed in a variety of fields such as linguistics, translation studies, natural language processing, etc., and part-of-speech tagging is one of the most common forms of corpus annotation. This master thesis presents an evaluation of three part-of-speech taggers for French. All systems are freely available for non-commercial use but differ in the approach to POS tagging as well as in the way they interface to the user. Two series of experiments are carried out where taggers are tested without training or tuning. The aim is to provide users with an overview of different alternatives for the morphosyntactic annotation of French corpora and the opportunity to choose the POS tagger that best suits their needs − whether it is in terms of the quality of the annotation with respect to the text typology, the format of the files to be processed or the skills required to deploy it.

Keywords
  • Part-of-speech tagging
  • French
  • Evaluation
  • User-oriented
  • Accuracy
  • MElt
  • TreeTagger
  • UDPipe 2.0
  • NLP
  • Corpus linguistics
  • Analyseurs morpho-syntaxiques
  • Français
  • Évaluation
  • Orientée-utilisateurs
  • Précision
  • TALN
  • Linguistique de corpus
Citation (ISO format)
MATTIUZZI, Silvia. An evaluation of part-of-speech taggers for French. Master, 2021.
Main files (1)
Master thesis
accessLevelPublic
Identifiers
  • PID : unige:156542
355views
341downloads

Technical informations

Creation11/03/2021 9:55:00 AM
First validation11/03/2021 9:55:00 AM
Update time03/28/2024 7:24:06 AM
Status update03/28/2024 7:24:06 AM
Last indexation12/17/2024 3:37:41 PM
All rights reserved by Archive ouverte UNIGE and the University of GenevaunigeBlack