Optimum SOM Distribution in Metric Spaces

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F60461373%3A22340%2F02%3A00007063" target="_blank" >RIV/60461373:22340/02:00007063 - isvavai.cz</a>

Result on the web

—

DOI - Digital Object Identifier

—

Result language

angličtina

Original language name

Optimum SOM Distribution in Metric Spaces

Original language description

Optimum SOM distribution is a combinatorial method of self-organization. The input patterns are supposed to be elements of metric space M with given distance measure. Let G be given undirected graph as a model of output metric space with node distance. The optimum SOM is a distribution of input patterns into graph nodes satisfying distance proportionality. There are various approaches to objective function design. The traditional penalty approach related to cluster analysis is compared with fuzzy logicfunction minimization. The discrete optimization task is solved as an NP-complete problem using combinatorial and stochastic global optimization techniques. Both traditional and fuzzy approaches are compared on illustrative example. The optimum SOM distribution is a general tool for the pattern categorization including patterns of observable states or outputs of technological process.

Czech name

—

Czech description

—

Type

D - Article in proceedings

CEP classification

JD - Use of computers, robotics and its application

OECD FORD branch

—

Project

—

Continuities

Z - Vyzkumny zamer (s odkazem do CEZ)

Publication year

2002

Confidentiality

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection

Proc. of 8th Int. Conference on Soft Computing MENDEL 2002

ISBN

80-214-2135-5

ISSN

—

e-ISSN

—

Number of pages

9

Pages from-to

287-291

Publisher name

VUT Brno

Place of publication

Brno

Event location

80-214-2135-5

Event date

Jun 5, 2002

Type of event by nationality

EUR - Evropská akce

UT code for WoS article

—