Catalogo Articoli (Spogli Riviste)

OPAC HELP

Titolo:
MLP based linear feature extraction for nonlinearly separable data
Autore:
Raudys, A; Long, JA;
Indirizzi:
S Bank Univ, Sch Comp Informat Syst & Math, London SE1 0AA, England S BankUniv London England SE1 0AA Syst & Math, London SE1 0AA, England
Titolo Testata:
PATTERN ANALYSIS AND APPLICATIONS
fascicolo: 4, volume: 4, anno: 2001,
pagine: 227 - 234
SICI:
1433-7541(2001)4:4<227:MBLFEF>2.0.ZU;2-V
Fonte:
ISI
Lingua:
ENG
Soggetto:
NEURAL-NETWORK; PATTERN;
Keywords:
feature extraction; linear; mapping; multilayer perceptron; principal components; sammon; transformation; visualisation;
Tipo documento:
Article
Natura:
Periodico
Settore Disciplinare:
Engineering, Computing & Technology
Citazioni:
22
Recensione:
Indirizzi per estratti:
Indirizzo: Raudys, A S Bank Univ, Sch Comp Informat Syst & Math, 103 Borough Rd, London SE1 0AA, England S Bank Univ 103 Borough Rd London England SE1 0AA 0AA, England
Citazione:
A. Raudys e J.A. Long, "MLP based linear feature extraction for nonlinearly separable data", PATTERN A A, 4(4), 2001, pp. 227-234

Abstract

A novel approach to linear feature extraction is presented. Most supervised feature extraction algorithms use mean square error or other measures based on the difference between expected and actual output values as a performance criterion. The novel approach presented here uses data visualisation together with an empirical classification error (percentage of cases classified incorrectly) as performance criterion. To find the optimal data transformation weights, the Multilayer Perceptron cost function with a special regularisation term is applied. The technique proposed is verified and compared with five competing mapping techniques with respect to visualisation and different classification error criteria. For comparison, two artificial and12 real world data sets are used.

ASDD Area Sistemi Dipartimentali e Documentali, Università di Bologna, Catalogo delle riviste ed altri periodici
Documento generato il 03/04/20 alle ore 19:49:11