An Evaluation of Self-Organizing Map Networks as a Robust Alternative to Factor Analysis in Data Mining Applications

Melody Y. Kiang, Ajith Kumar

Research output: Contribution to journalArticle

35 Citations (Scopus)

Abstract

Kohonen's self-organizing map (SOM) network is one of the most important network architectures developed during the 1980s. The main function of SOM networks is to map the input data from an n-dimensional space to a lower dimensional (usually one- or two-dimensional) plot while maintaining the original topological relations. Therefore, it can be viewed as an analog of factor analysis. In this research, we evaluate the feasibility of using SOM networks as a robust alternative to factor analysis and clustering for data mining applications. Specifically, we compare SOM network solutions to factor analytic and K-Means clustering solutions on simulated data sets with known underlying factor and cluster structures. The comparisons indicate that the SOM networks provide solutions superior to unrotated factor solutions in general and provide more accurate recovery of underlying cluster structures when the input data are skewed. Our findings suggest that SOM networks can provide robust alternatives to traditional factor analysis and clustering techniques in data mining applications.

Original languageEnglish (US)
Pages (from-to)177-194
Number of pages18
JournalInformation Systems Research
Volume12
Issue number2
StatePublished - Jun 2001
Externally publishedYes

Fingerprint

Self organizing maps
Factor analysis
Data mining
factor analysis
evaluation
Network architecture
Self-organizing map
Evaluation
Recovery
Factors

Keywords

  • Clustering Analysis
  • Data Mining
  • Data Reductive
  • Factor Analysis
  • Kohonen Networks

ASJC Scopus subject areas

  • Library and Information Sciences

Cite this

An Evaluation of Self-Organizing Map Networks as a Robust Alternative to Factor Analysis in Data Mining Applications. / Kiang, Melody Y.; Kumar, Ajith.

In: Information Systems Research, Vol. 12, No. 2, 06.2001, p. 177-194.

Research output: Contribution to journalArticle

@article{b5a081f9ab14494c803fe2a032c89ce9,
title = "An Evaluation of Self-Organizing Map Networks as a Robust Alternative to Factor Analysis in Data Mining Applications",
abstract = "Kohonen's self-organizing map (SOM) network is one of the most important network architectures developed during the 1980s. The main function of SOM networks is to map the input data from an n-dimensional space to a lower dimensional (usually one- or two-dimensional) plot while maintaining the original topological relations. Therefore, it can be viewed as an analog of factor analysis. In this research, we evaluate the feasibility of using SOM networks as a robust alternative to factor analysis and clustering for data mining applications. Specifically, we compare SOM network solutions to factor analytic and K-Means clustering solutions on simulated data sets with known underlying factor and cluster structures. The comparisons indicate that the SOM networks provide solutions superior to unrotated factor solutions in general and provide more accurate recovery of underlying cluster structures when the input data are skewed. Our findings suggest that SOM networks can provide robust alternatives to traditional factor analysis and clustering techniques in data mining applications.",
keywords = "Clustering Analysis, Data Mining, Data Reductive, Factor Analysis, Kohonen Networks",
author = "Kiang, {Melody Y.} and Ajith Kumar",
year = "2001",
month = "6",
language = "English (US)",
volume = "12",
pages = "177--194",
journal = "Information Systems Research",
issn = "1047-7047",
publisher = "INFORMS Inst.for Operations Res.and the Management Sciences",
number = "2",

}

TY - JOUR

T1 - An Evaluation of Self-Organizing Map Networks as a Robust Alternative to Factor Analysis in Data Mining Applications

AU - Kiang, Melody Y.

AU - Kumar, Ajith

PY - 2001/6

Y1 - 2001/6

N2 - Kohonen's self-organizing map (SOM) network is one of the most important network architectures developed during the 1980s. The main function of SOM networks is to map the input data from an n-dimensional space to a lower dimensional (usually one- or two-dimensional) plot while maintaining the original topological relations. Therefore, it can be viewed as an analog of factor analysis. In this research, we evaluate the feasibility of using SOM networks as a robust alternative to factor analysis and clustering for data mining applications. Specifically, we compare SOM network solutions to factor analytic and K-Means clustering solutions on simulated data sets with known underlying factor and cluster structures. The comparisons indicate that the SOM networks provide solutions superior to unrotated factor solutions in general and provide more accurate recovery of underlying cluster structures when the input data are skewed. Our findings suggest that SOM networks can provide robust alternatives to traditional factor analysis and clustering techniques in data mining applications.

AB - Kohonen's self-organizing map (SOM) network is one of the most important network architectures developed during the 1980s. The main function of SOM networks is to map the input data from an n-dimensional space to a lower dimensional (usually one- or two-dimensional) plot while maintaining the original topological relations. Therefore, it can be viewed as an analog of factor analysis. In this research, we evaluate the feasibility of using SOM networks as a robust alternative to factor analysis and clustering for data mining applications. Specifically, we compare SOM network solutions to factor analytic and K-Means clustering solutions on simulated data sets with known underlying factor and cluster structures. The comparisons indicate that the SOM networks provide solutions superior to unrotated factor solutions in general and provide more accurate recovery of underlying cluster structures when the input data are skewed. Our findings suggest that SOM networks can provide robust alternatives to traditional factor analysis and clustering techniques in data mining applications.

KW - Clustering Analysis

KW - Data Mining

KW - Data Reductive

KW - Factor Analysis

KW - Kohonen Networks

UR - http://www.scopus.com/inward/record.url?scp=0035588333&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0035588333&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:0035588333

VL - 12

SP - 177

EP - 194

JO - Information Systems Research

JF - Information Systems Research

SN - 1047-7047

IS - 2

ER -