Mixtures of factor analyzers are becoming more and more popular in the area of model based clustering of high-dimensional data. In data modeling, according to the likelihood approach, it is well known that the loglikelihood function may present spurious maxima and singularities and this is due to specific patterns of the estimated covariance structure. To reduce such drawbacks, we propose a constrained procedure which maintains the eigenvalues of the covariance structure into suitable ranges. Applications of this approach in robust clustering is then outlined
Greselin, F., Ingrassia, S. (2012). Constrained EM algorithms for Gaussian mixtures of factor analyzers. In A. Colubi, C. Croux, E.J. Kontoghiorghes, H.K. Van Dijk (a cura di), Proceedings of the 5th International Conference of the ERCIM WG on Computing & Statistics (ERCIM'12) (pp. 34-34). ERCIM WG on Computing & Statistics.
Constrained EM algorithms for Gaussian mixtures of factor analyzers
Greselin, F;
2012
Abstract
Mixtures of factor analyzers are becoming more and more popular in the area of model based clustering of high-dimensional data. In data modeling, according to the likelihood approach, it is well known that the loglikelihood function may present spurious maxima and singularities and this is due to specific patterns of the estimated covariance structure. To reduce such drawbacks, we propose a constrained procedure which maintains the eigenvalues of the covariance structure into suitable ranges. Applications of this approach in robust clustering is then outlinedFile | Dimensione | Formato | |
---|---|---|---|
GI Ercim 2012 Constrained EM algorithms for Gaussian mixtures of factor analyzers.pdf
accesso aperto
Dimensione
201.39 kB
Formato
Adobe PDF
|
201.39 kB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.