Skip to Main content Skip to Navigation
Book sections

Recoder les variables pour obtenir un modèle implicatif optimal

Martine Cadot 1
1 MULTISPEECH - Speech Modeling for Facilitating Oral-Based Communication
Inria Nancy - Grand Est, LORIA - NLPKD - Department of Natural Language Processing & Knowledge Discovery
Abstract : A number of methods are available for deriving a categorization model of type XY out of a set of individual data, where X is a set of individual numerical features and Y their categories. We develop a brief overview of these methods by making use of the most popular ones for processing the well-known "Fisher’s Iris" dataset. The comparison of the resulting models encourages us to give preference to ISA (Implicative Statistical Analysis) for this specific type of data, on condition of a thorough recoding of the quantitative variables. This paper incorporates and expands a communication made during A.S.I.8 conference (Cadot et al. 2015) in which we show the interest of the chosen methodology (ISA after a specific recoding step) for the processing of acoustic data.
Document type :
Book sections
Complete list of metadata

Cited literature [15 references]  Display  Hide  Download
Contributor : Martine Cadot <>
Submitted on : Wednesday, November 16, 2016 - 10:39:53 PM
Last modification on : Thursday, March 5, 2020 - 4:55:05 PM
Long-term archiving on: : Thursday, March 16, 2017 - 3:47:11 PM


Files produced by the author(s)


  • HAL Id : hal-01398229, version 1



Martine Cadot. Recoder les variables pour obtenir un modèle implicatif optimal. Régis Gras. L'Analyse Statisqtique Implicative, Cépaduès, 2016. ⟨hal-01398229⟩



Record views


Files downloads