Kohonen SOM Learning Strategy and Country Classification

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21340%2F17%3A00316574" target="_blank" >RIV/68407700:21340/17:00316574 - isvavai.cz</a>

Výsledek na webu

—

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Kohonen SOM Learning Strategy and Country Classification

Popis výsledku v původním jazyce

The Self-Organized Mapping (SOM) represents a traditional tool for multidimen- sional data analysis overperforming analytical power of cluster analysis. But there are possible difficulties when the SOM is applied to data patterns of large size. We present testing example using iris dataset. Our approach is mainly used for macro-economical data analysis which is based on logarithmic differences, pattern dimensionality reduction and finalization of data analysis using Kohonen SOM learning. General methodology was applied to main economic indicators describing the situation of thirty five countries during more than twenty years. The used dataset comes from regularly published statistics of European Commission. The main aim is to identify the similarities of countries. The role of SOM topology, learning strategy and reduced pattern size can be also used to predict behaviour during crisis based on the identified similarity and known.

Název v anglickém jazyce

Kohonen SOM Learning Strategy and Country Classification

Popis výsledku anglicky

The Self-Organized Mapping (SOM) represents a traditional tool for multidimen- sional data analysis overperforming analytical power of cluster analysis. But there are possible difficulties when the SOM is applied to data patterns of large size. We present testing example using iris dataset. Our approach is mainly used for macro-economical data analysis which is based on logarithmic differences, pattern dimensionality reduction and finalization of data analysis using Kohonen SOM learning. General methodology was applied to main economic indicators describing the situation of thirty five countries during more than twenty years. The used dataset comes from regularly published statistics of European Commission. The main aim is to identify the similarities of countries. The role of SOM topology, learning strategy and reduced pattern size can be also used to predict behaviour during crisis based on the identified similarity and known.

Druh

O - Ostatní výsledky

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach<br>I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Rok uplatnění

2017

Kód důvěrnosti údajů

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