Catalogo Articoli (Spogli Riviste)

OPAC HELP

Titolo:
SMEM algorithm for mixture regression problems
Autore:
Suzuki, S; Ueda, N;
Indirizzi:
NTT, Commun Sci Labs, Kyoto 6190237, Japan NTT Kyoto Japan 6190237NTT, Commun Sci Labs, Kyoto 6190237, Japan
Titolo Testata:
ELECTRONICS AND COMMUNICATIONS IN JAPAN PART II-ELECTRONICS
fascicolo: 12, volume: 84, anno: 2001,
pagine: 54 - 62
SICI:
8756-663X(2001)84:12<54:SAFMRP>2.0.ZU;2-F
Fonte:
ISI
Lingua:
ENG
Keywords:
EM algorithm; regression; NGnet; mixture model; maximum likelihood estimation; joint distribution;
Tipo documento:
Article
Natura:
Periodico
Settore Disciplinare:
Engineering, Computing & Technology
--discip_EC--
Citazioni:
5
Recensione:
Indirizzi per estratti:
Indirizzo: Suzuki, S NTT, Commun Sci Labs, Kyoto 6190237, Japan NTT Kyoto Japan 6190237 Commun Sci Labs, Kyoto 6190237, Japan
Citazione:
S. Suzuki e N. Ueda, "SMEM algorithm for mixture regression problems", ELEC C JP 2, 84(12), 2001, pp. 54-62

Abstract

We investigate applications of the SMEM (Split and Merge EM) algorithm to regression problems. We use the NGnet (Normalized Gaussian network) as a mixture regression model in which a set of input and output variables is dealt with as a joint probability. We describe the split and merge operations and criteria necessary for applying the SMEM algorithm to the NGnet. We showthe usefulness of the SMEM algorithm for regression problems through experiments using artificial and real data. (C) 2001 Scripta Technica.

ASDD Area Sistemi Dipartimentali e Documentali, Università di Bologna, Catalogo delle riviste ed altri periodici
Documento generato il 22/01/20 alle ore 12:55:55