Modeling Comparisons for some Classification Methods, Bayesian, Neural and Traditional Cluster Techniques

  1. Lévy Mangin, Jean| Pierre 1
  2. Moriano, Juan Antonio 2
  3. Bourgault, Normand 1
  1. 1 Université du Québec en Outaouais, Québec
  2. 2 Faculty of Psychology, Universidad Nacional de Educación a Distancia, Madrid
Journal:
CIENCIA ergo-sum

ISSN: 1405-0269

Year of publication: 2010

Volume: 17

Issue: 2

Pages: 127-135

Type: Article

DIALNET GOOGLE SCHOLAR lock_openDialnet editor

More publications in: CIENCIA ergo-sum

Abstract

This article compares some classification methods that would be very useful for clustering purposes mainly in marketing. First of them are based on Latent Class Mixture Modeling with training data and without training data. The second set of techniques is based on Neural Networks Classification Method and finally we will present methods based on more classical techniques like K-Means and Hierarchical Cluster Analysis techniques.

Data source: Dialnet