Porovnání schopností různých druhů neuronových sítí pro účely shlukové analýzy

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F49777513%3A23220%2F09%3A00504919" target="_blank" >RIV/49777513:23220/09:00504919 - isvavai.cz</a>

Výsledek na webu

—

DOI - Digital Object Identifier

—

Jazyk výsledku

čeština

Název v původním jazyce

Porovnání schopností různých druhů neuronových sítí pro účely shlukové analýzy

Popis výsledku v původním jazyce

For cluster analysis purpose we can use several types of artificial neural networks. We can choose self-organizing networks or networks with teacher. Neural networks with teacher need the data set with correct pre-classification for the training phase. After the training process these networks are able to classify by teaching class. Self-organizing networks (known as the networks with competitive learning) are able to self-classify and make clusters, identifying clustering data. This principle is in fact similar to classic statistical cluster analysis method. This article is focused to all kinds of neural networks, which are proper for purpose of cluster analysis (multi-layer perceptron networks, RBF networks, linear networks and networks with competitive learning) and appreciate the possibilities for use to both cluster analysis methods.

Název v anglickém jazyce

The comparison of neural networks abilities for cluster analysis purpose

Popis výsledku anglicky

For cluster analysis purpose we can use several types of artificial neural networks. We can choose self-organizing networks or networks with teacher. Neural networks with teacher need the data set with correct pre-classification for the training phase. After the training process these networks are able to classify by teaching class. Self-organizing networks (known as the networks with competitive learning) are able to self-classify and make clusters, identifying clustering data. This principle is in fact similar to classic statistical cluster analysis method. This article is focused to all kinds of neural networks, which are proper for purpose of cluster analysis (multi-layer perceptron networks, RBF networks, linear networks and networks with competitive learning) and appreciate the possibilities for use to both cluster analysis methods.

Druh

O - Ostatní výsledky

CEP obor

JB - Senzory, čidla, měření a regulace

OECD FORD obor

—

Projekt

—

Návaznosti

Z - Vyzkumny zamer (s odkazem do CEZ)

Rok uplatnění

2009

Kód důvěrnosti údajů

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