Parameter Setting for Evolutionary Latent Class Clustering

Damien Tessier 1 Marc Schoenauer 1 Christophe Biernacki 2 Gilles Celeux 3 Gérard Govaert 4
1 TANC - Algorithmic number theory for cryptology
Inria Saclay - Ile de France, LIX - Laboratoire d'informatique de l'École polytechnique [Palaiseau]
3 SELECT - Model selection in statistical learning
Inria Saclay - Ile de France, LMO - Laboratoire de Mathématiques d'Orsay
Abstract : The latent class model or multivariate multinomial mixture is a powerful model for clustering discrete data. This model is expected to be useful to represent non-homogeneous populations. It uses a conditional independence assumption given the latent class to which a statistical unit is belonging. However, it leads to a criterion that proves difficult to optimise by the standard approach based on the EM algorithm. An Evolutionary Algorithms is designed to tackle this discrete optimisation problem, and an extensive parameter study on a large artificial dataset allows to derive stable parameters. Those parameters are then validated on other artificial datasets, as well as on some well-known real data: the Evolutionary Algorithm performs repeatedly better than other standard clustering techniques on the same data.
Complete list of metadatas

Cited literature [13 references]  Display  Hide  Download

https://hal.inria.fr/inria-00179186
Contributor : Marc Schoenauer <>
Submitted on : Sunday, October 14, 2007 - 7:39:09 AM
Last modification on : Monday, February 10, 2020 - 6:13:44 PM
Long-term archiving on: Sunday, April 11, 2010 - 10:59:13 PM

File

latentEA.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : inria-00179186, version 1

Citation

Damien Tessier, Marc Schoenauer, Christophe Biernacki, Gilles Celeux, Gérard Govaert. Parameter Setting for Evolutionary Latent Class Clustering. Second International Symposium, ISICA 2007, Sep 2007, Wuhan, China. pp.472-484. ⟨inria-00179186⟩

Share

Metrics

Record views

726

Files downloads

899