Representing interaction in multiway contingency tables: MIDOVA, CA and log-linear model

Martine Cadot 1 Alain Lelu 2, 3, 4
1 ABC - Machine Learning and Computational Biology
LORIA - Laboratoire Lorrain de Recherche en Informatique et ses Applications
2 KIWI - Knowledge Information and Web Intelligence
LORIA - Laboratoire Lorrain de Recherche en Informatique et ses Applications
Abstract : Beside CA and log-linear model, issued from the statistics domain, other research streams originating in Artificial Intelligence have coped with the interacting variables problem: we will present here the extension to categorical variables of our results on extracting and statistically validating " itemsets " in boolean datatables. We coined MIDOVA (Multidimensional Interaction Differential of Variation) our method for highlighting and representing complex links between qualitative variables, which includes interaction, well-suited to socio-economic data. We will compare it to the CA and log-linear model approaches, using the same 3-way example as Escofier and her colleagues. We will show that out method is effective for general N-way interactions (N may be far greater than 3), whether symmetrically or not, and results both in easy and detailed interpretability, as CA does, and in statistical significance testing, as the log-linear model does in the case of few variables.
Type de document :
Communication dans un congrès
Jörg Blasius, and Michael Greenacre, and Jérôme Pagès. 6th International Conference on Correspondence Analysis and Related Methods - CARME 2011, Feb 2011, Rennes, France. 2010, 〈http://carme2011.agrocampus-ouest.fr/book_of_abstracts/Cadot_Lelu.pdf〉
Liste complète des métadonnées

Littérature citée [7 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/inria-00547886
Contributeur : Alain Lelu <>
Soumis le : jeudi 17 mars 2011 - 07:00:02
Dernière modification le : mardi 24 avril 2018 - 13:34:58
Document(s) archivé(s) le : samedi 18 juin 2011 - 02:20:08

Fichier

Cadot_Lelu.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : inria-00547886, version 1

Collections

Citation

Martine Cadot, Alain Lelu. Representing interaction in multiway contingency tables: MIDOVA, CA and log-linear model. Jörg Blasius, and Michael Greenacre, and Jérôme Pagès. 6th International Conference on Correspondence Analysis and Related Methods - CARME 2011, Feb 2011, Rennes, France. 2010, 〈http://carme2011.agrocampus-ouest.fr/book_of_abstracts/Cadot_Lelu.pdf〉. 〈inria-00547886〉

Partager

Métriques

Consultations de la notice

312

Téléchargements de fichiers

153