Publication details

Minimum Information Loss Cluster Analysis for Cathegorical Data

Journal Article

Grim Jiří, Hora Jan

serial: Lecture Notes in Computer Science vol.2007, p. 233-247

action: International Conference on Machine Learning and Data Mining MLDM 2007 /5./, (Leipzig, DE, 18.07.2007-20.07.2007)

research: CEZ:AV0Z10750506

project(s): 1M0572, GA MŠk, GA102/07/1594, GA ČR, 2C06019, GA MŠk

keywords: Cluster Analysis, Cathegorical Data, EM algorithm

abstract (eng):

The EM algorithm has been used repeatedly to identify latent classes in categorical data by estimating finite distribution mixtures of produkt components. Unfortunately, the underlying mixtures are not uniquely identifiable and, moreover, the estimated mixture parameters are starting-point dependent. For this reason we use the latent class model only to define a set of ``elementary'' classes by estimating a mixture of a large number components. We propose a hierarchical ``bottom up'' cluster analysis based on unifying the elementary latent classes sequentially. The clustering procedure is controlled by minimum information loss criterion.

abstract (cze):

Shluková analýza kategoriálních dat s využitím kriteria minimální ztráty informace.