Enhancing the selection of a model-based clustering with external qualitative variables

Jean-Patrick Baudry 1, * Margarida Cardoso 2 Gilles Celeux 3 Maria-José Amorim 4 Ana Sousa Ferreira 2
* Corresponding author
2 BRU-UNIDE
IST - Instituto Superior Técnico, Universidade Técnica de Lisboa
3 SELECT - Model selection in statistical learning
Inria Saclay - Ile de France, LMO - Laboratoire de Mathématiques d'Orsay
4 ISEL
IST - Instituto Superior Técnico, Universidade Técnica de Lisboa
Abstract : In cluster analysis, it could be useful to interpret the obtained partition with respect to external qualitative variables. An approach is proposed in the model-based clustering context to select a model and a number of clusters in order to get a partition which both provides a good fit with the data and is well related to the external variables. This approach makes use of the integrated joint likelihood of the data and the partitions at hand, namely the model-based partition and the partitions associated to the external variables. It is worth noticing that the known partitions are only used to select a relevant mixture model. Each mixture model is fitted by the maximum likelihood methodology from the data. Numerical experiments illustrate the promising behaviour of the derived criterion.
Document type :
Preprints, Working Papers, ...
Complete list of metadatas

Cited literature [11 references]  Display  Hide  Download

https://hal.sorbonne-universite.fr/hal-00747854
Contributor : Jean-Patrick Baudry <>
Submitted on : Friday, November 2, 2012 - 12:15:28 PM
Last modification on : Monday, January 13, 2020 - 1:59:27 PM
Long-term archiving on: Saturday, December 17, 2016 - 7:02:02 AM

File

HAL.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00747854, version 1

Citation

Jean-Patrick Baudry, Margarida Cardoso, Gilles Celeux, Maria-José Amorim, Ana Sousa Ferreira. Enhancing the selection of a model-based clustering with external qualitative variables. 2012. ⟨hal-00747854⟩

Share

Metrics

Record views

1380

Files downloads

888