MLP AND DEFAULT ARTMAP NEURAL NETWORKS FOR BUSINESS CLASSIFICATION

Anatoli Nachev

MLP AND DEFAULT ARTMAP NEURAL NETWORKS FOR BUSINESS CLASSIFICATION

Anatoli Nachev

Business Information Systems

Research output: Chapter in Book or Conference Publication/Proceeding › Conference Publication › peer-review

Abstract

This study shows in an empirical form that a specific class of neural networks, Default ARTMAP, is superior to the most common MLP neural networks in predicting business insolvency Our experiments are based on data collected from 88 Irish firms over six years We experimented with different network parameters, variable selection, and daunt preprocessing Our results were verified and validated by several performance metrics, ROC analysis and leave-one-out cross-validation

Original language	English (Ireland)
Title of host publication	INTELLIGENT DECISION MAKING SYSTEMS
Publisher	WORLD SCIENTIFIC AND ENGINEERING ACAD AND SOC
Number of pages	5
Volume	2
Publication status	Published - 1 Jan 2010

Authors (Note for portal: view the doc link for the full list of authors)

Authors
Nachev, A;Stoyanov, B

Cite this

@inproceedings{79071cd0f651466c8aacaf1269e81439,

title = "MLP AND DEFAULT ARTMAP NEURAL NETWORKS FOR BUSINESS CLASSIFICATION",

abstract = "This study shows in an empirical form that a specific class of neural networks, Default ARTMAP, is superior to the most common MLP neural networks in predicting business insolvency Our experiments are based on data collected from 88 Irish firms over six years We experimented with different network parameters, variable selection, and daunt preprocessing Our results were verified and validated by several performance metrics, ROC analysis and leave-one-out cross-validation",

author = "Anatoli Nachev",

year = "2010",

month = jan,

day = "1",

language = "English (Ireland)",

volume = "2",

booktitle = "INTELLIGENT DECISION MAKING SYSTEMS",

publisher = "WORLD SCIENTIFIC AND ENGINEERING ACAD AND SOC",

}

TY - GEN

T1 - MLP AND DEFAULT ARTMAP NEURAL NETWORKS FOR BUSINESS CLASSIFICATION

AU - Nachev, Anatoli

PY - 2010/1/1

Y1 - 2010/1/1

N2 - This study shows in an empirical form that a specific class of neural networks, Default ARTMAP, is superior to the most common MLP neural networks in predicting business insolvency Our experiments are based on data collected from 88 Irish firms over six years We experimented with different network parameters, variable selection, and daunt preprocessing Our results were verified and validated by several performance metrics, ROC analysis and leave-one-out cross-validation

AB - This study shows in an empirical form that a specific class of neural networks, Default ARTMAP, is superior to the most common MLP neural networks in predicting business insolvency Our experiments are based on data collected from 88 Irish firms over six years We experimented with different network parameters, variable selection, and daunt preprocessing Our results were verified and validated by several performance metrics, ROC analysis and leave-one-out cross-validation

M3 - Conference Publication

VL - 2

BT - INTELLIGENT DECISION MAKING SYSTEMS

PB - WORLD SCIENTIFIC AND ENGINEERING ACAD AND SOC

ER -

MLP AND DEFAULT ARTMAP NEURAL NETWORKS FOR BUSINESS CLASSIFICATION

Abstract

Authors (Note for portal: view the doc link for the full list of authors)

Fingerprint

Cite this