"Praha" . . "P(1M0572), P(GA102/07/1594), Z(AV0Z10750506), Z(MSM6840770039)" . . "[654E91F42C26]" . "The EM algorithm has been used repeatedly to identify latent classes in categorical data by estimating finite distribution mixtures of product 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. As such a mixture we use also an optimally smoothed kernel estimate. 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."@en . . . . "RIV/67985556:_____/07:00098540" . . . "\u010Cesk\u00E1 technika - nakladatelstv\u00ED \u010CVUT" . "Shlukov\u00E1n\u00ED kategori\u00E1ln\u00EDch dat je \u010Dasto \u0159e\u0161eno hled\u00E1n\u00EDm tzv. latentn\u00EDch t\u0159\u00EDd pomoc\u00ED EM algoritmu. Tento p\u0159\u00EDstup ov\u0161em z\u00E1vis\u00ED na po\u010D\u00E1te\u010Dn\u00EDm \u0159e\u0161en\u00ED a nar\u00E1\u017E\u00ED na probl\u00E9m neidentifikovatelosti sm\u011Bsi. Popisovan\u00E1 metoda vyhled\u00E1v\u00E1 shluky nikoliv jako jednotliv\u00E9 komponenty sm\u011Bsi jako v p\u0159\u00EDpad\u011B latentn\u00EDch t\u0159\u00EDd, ale jako podsm\u011Bsi vznikl\u00E9 slou\u010Den\u00EDm n\u011Bkolika jednoduch\u00FDch t\u0159\u00EDd z odhadnut\u00E9 distribu\u010Dn\u00ED sm\u011Bsi s vy\u0161\u0161\u00EDm po\u010Dtem komponent. Extr\u00E9mn\u00ED variantou takov\u00E9 sm\u011Bsi m\u016F\u017Ee b\u00FDt j\u00E1drov\u00FD odhad, jeho\u017E optim\u00E1ln\u00ED vyhlazen\u00ED je v pr\u00E1ci pops\u00E1no. V pr\u00E1ci je d\u00E1le p\u0159edstavena metoda hierarchick\u00E9ho shlukov\u00E1n\u00ED s krit\u00E9riem nejmen\u0161\u00ED informa\u010Dn\u00ED ztr\u00E1ty."@cs . "EM algorithm; distribution mixtures; cluster analysis; cathegorial data"@en . "Hora, Jan" . "10"^^ . "Doktorandsk\u00E9 dny 2007" . . . "Informational Cathegorical Data Clustering"@en . "57;66" . . . "Informational Cathegorical Data Clustering" . "RIV/67985556:_____/07:00098540!RIV08-AV0-67985556" . "2007-11-16+01:00"^^ . . . . "Informational Cathegorical Data Clustering" . "426614" . "Informational Cathegorical Data Clustering"@en . "Informa\u010Dn\u00ED shlukov\u00E1n\u00ED kategori\u00E1ln\u00EDch dat"@cs . . "978-80-01-03913-7" . "1"^^ . . . "1"^^ . "The EM algorithm has been used repeatedly to identify latent classes in categorical data by estimating finite distribution mixtures of product 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. As such a mixture we use also an optimally smoothed kernel estimate. 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." . . . . "Praha" . "Informa\u010Dn\u00ED shlukov\u00E1n\u00ED kategori\u00E1ln\u00EDch dat"@cs . .