ESB neural-networks-based classifiers

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F61988987%3A17310%2F12%3AA13017K2" target="_blank" >RIV/61988987:17310/12:A13017K2 - isvavai.cz</a>

Result on the web

<a href="http://projects.osu.eu/intsys/software/esb_nnclass.html" target="_blank" >http://projects.osu.eu/intsys/software/esb_nnclass.html</a>

DOI - Digital Object Identifier

—

Result language

angličtina

Original language name

ESB neural-networks-based classifiers

Original language description

In this software we demonstrate an experimental study which was aimed to explore the possibility of using neural network as the base algorithms for so called weak classifiers. In our experiment neural networks went only through one adaptation cycle. We have generated experimental ensembles of 100 weak classifiers based on five types of neural networks. A total of 3,240 of such ensembles have been created and tested. We have proposed filtering of input as the new diversity-achieving method. We have tested this method in common with two others methods. The experiment has been conducted over the MNIST database.

Czech name

—

Czech description

—

Type

R - Software

CEP classification

IN - Informatics

OECD FORD branch

—

Project

—

Continuities

S - Specificky vyzkum na vysokych skolach

Publication year

2012

Confidentiality

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

Internal product ID

NNCLASS

Technical parameters

Michal Janošek, OU PřF KIP, michal.janosek@osu.cz

Economical parameters

weak classifiers - diversity enhancement

Owner IČO

61988987

Owner name

Ostravská Univerzita v Ostravě