Clustering categorical functional data ...
Document type :
Autre communication scientifique (congrès sans actes - poster - séminaire...): Communication dans un congrès sans actes
Permalink :
Title :
Clustering categorical functional data Application to medical discharge letters
Author(s) :
Vandewalle, Vincent [Auteur]
Evaluation des technologies de santé et des pratiques médicales - ULR 2694 [METRICS]
METRICS : Evaluation des technologies de santé et des pratiques médicales - ULR 2694
Cozma, Cristina [Auteur]
Preda, Cristian [Auteur]
Laboratoire Paul Painlevé - UMR 8524 [LPP]
Laboratoire Paul Painlevé - UMR 8524
Evaluation des technologies de santé et des pratiques médicales - ULR 2694 [METRICS]
METRICS : Evaluation des technologies de santé et des pratiques médicales - ULR 2694
Cozma, Cristina [Auteur]
Preda, Cristian [Auteur]
Laboratoire Paul Painlevé - UMR 8524 [LPP]
Laboratoire Paul Painlevé - UMR 8524
Conference title :
8th International Conference of the ERCIM WG on Computational and Methodological Statistics
City :
Londres
Country :
France
Start date of the conference :
2015-12-12
Publication date :
2015-12-12
HAL domain(s) :
Mathématiques [math]
English abstract : [en]
Categorical functional data represented by paths of a stochastic jump process are considered for clustering. For paths of the same length, the extension of the multiple correspondence analysis allows the use of well-known ...
Show more >Categorical functional data represented by paths of a stochastic jump process are considered for clustering. For paths of the same length, the extension of the multiple correspondence analysis allows the use of well-known methods for clustering finite dimensional data. When the paths are of different lengths, the analysis is more complex. In this case, for Markov models we propose an EM algorithm to estimate a mixture of Markov processes. A simulation study as well as a real application on hospital stays will be presented.Show less >
Show more >Categorical functional data represented by paths of a stochastic jump process are considered for clustering. For paths of the same length, the extension of the multiple correspondence analysis allows the use of well-known methods for clustering finite dimensional data. When the paths are of different lengths, the analysis is more complex. In this case, for Markov models we propose an EM algorithm to estimate a mixture of Markov processes. A simulation study as well as a real application on hospital stays will be presented.Show less >
Language :
Anglais
Audience :
Internationale
Popular science :
Non
Administrative institution(s) :
CHU Lille
CNRS
Université de Lille
CNRS
Université de Lille
Submission date :
2020-06-08T14:10:09Z