Co-clustering de courbes fonctionnelles multivariées

Abstract : The exponential growth of smart devices in all aspect of everyday life, leads to the collection of high frequency data for a same individual. Those devices also ease the collection of several variables simultaneously for an individual, which results in growing needs of methods to summarise and read such multivariate functional data. This work shows a new functional co-clustering method in order to help highlighting groups of individuals and variables that look alike. This method relies on a functional latent block model and model inference is done with a SEM-Gibbs algorithm. The model efficiency will be shown on a practical example of smart houses where the consumption of electricity and the temperature is monitored over time.
Document type :
Conference papers
Complete list of metadatas

Cited literature [5 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-02095004
Contributor : Amandine Schmutz <>
Submitted on : Wednesday, April 10, 2019 - 10:30:31 AM
Last modification on : Wednesday, May 1, 2019 - 1:36:28 AM

File

JDS2019.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02095004, version 1

Collections

Citation

Amandine Schmutz, Julien Jacques, Charles Bouveyron, Laurence Chèze, Pauline Martin. Co-clustering de courbes fonctionnelles multivariées. Journées des Statistiques, Jun 2019, Nancy, France. ⟨hal-02095004⟩

Share

Metrics

Record views

32

Files downloads

19