Identification de descripteurs pour la caractérisation de registres

Abstract : The paper presents a study of linguistic features for the characterization of a text according to its language register (formal, neutral, informal). This study aims at laying a first milestones for future work on this subject (e.g., classification, discriminating patterns extraction, etc.). From a state of the art conducted on the notion of register in linguistics and sociolinguistics, we have identified a list of 72 relevant descriptors. In this paper, we present the first 30 ones that we could validate on a corpus of French texts from distinct registers. MOTS-CLÉS : registres de langue, descripteur linguistique, validation.
Complete list of metadatas

Cited literature [20 references]  Display  Hide  Download

https://hal.inria.fr/hal-02002612
Contributor : Gwénolé Lecorvé <>
Submitted on : Thursday, January 31, 2019 - 6:09:09 PM
Last modification on : Thursday, February 7, 2019 - 2:57:38 PM
Long-term archiving on : Wednesday, May 1, 2019 - 7:11:36 PM

File

paper 9.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02002612, version 1

Citation

Jade Mekki, Delphine Battistelli, Gwénolé Lecorvé, Nicolas Béchet. Identification de descripteurs pour la caractérisation de registres. Rencontre des jeunes chercheurs en traitement automatique du langage naturel et recherche d'information (CORIA-TALN-RJC), May 2018, Rennes, France. ⟨hal-02002612⟩

Share

Metrics

Record views

43

Files downloads

52