Skip to Main content Skip to Navigation
Journal articles

La textométrie en question

Abstract : Considering that textometric statistical analysis of textual data is based on plain counts (which are just numeric values) of words (which are defined as rough character string tokens), this approach shows obvious limits from a linguistic point of view. Moreover, Textometry uses formal and mathematical models to analyse corpora, but results are often delivered in a very informal manner, without any quantified evaluation procedure like the ones that are applied in the Natural Language Processing field. The present paper aims at giving an in-depth understanding of the textometric methodology, so that such critical points may not be relevant anymore. Then, Textometry can meet the requirements of a linguistic-aware and scientific study of textual data.
Complete list of metadata

Cited literature [39 references]  Display  Hide  Download
Contributor : Bénédicte Pincemin Connect in order to contact the contributor
Submitted on : Friday, July 17, 2020 - 5:53:08 PM
Last modification on : Sunday, June 26, 2022 - 1:10:41 AM
Long-term archiving on: : Tuesday, December 1, 2020 - 12:38:34 AM


Files produced by the author(s)


Distributed under a Creative Commons Attribution - NonCommercial - ShareAlike 4.0 International License


  • HAL Id : halshs-02902088, version 1


Bénédicte Pincemin. La textométrie en question. Le Français Moderne - Revue de linguistique Française, CILF (conseil international de la langue française), 2020, Linguistique et traitements quantitatifs, 88 (1), pp.26-43. ⟨halshs-02902088⟩



Record views


Files downloads