Verifying the Proximity and Size Hypothesis for Self-Organizing Maps

Research output: Contribution to journalArticle

43 Citations (Scopus)

Abstract

The Kohonen Self-Organizing Map (SOM) is an unsupervised learning technique for summarizing high-dimensional data so that similar inputs are, in general, mapped close to one another. When applied to textual data, SOM has been shown to be able to group together related concepts in a data collection and to present major topics within the collection with larger regions. This article presents research in which we sought to validate these properties of SOM, called the Proximity and Size Hypotheses, through a user evaluation study. Building upon our previous research in automatic concept generation and classification, we demonstrated that the Kohonen SOM was able to perform concept clustering effectively, based on its concept precision and recall7 scores as judged by human experts. We also demonstrated a positive relationship between the size of an SOM region and the number of documents contained in the region. We believe this research has established the Kohonen SOM algorithm as an intuitively appealing and promising neural-network-based textual classification technique for addressing part of the longstanding "information overload" problem.

Original languageEnglish (US)
Pages (from-to)57-70
Number of pages14
JournalJournal of Management Information Systems
Volume16
Issue number3
StatePublished - Dec 1999

Fingerprint

Self organizing maps
neural network
Unsupervised learning
expert
Self-organizing map
Proximity
evaluation
learning
Neural networks
Group

Keywords

  • Document clustering techniques
  • Experimental research
  • Group support systems
  • Self-organizing maps
  • Unsupervised learning algorithms

ASJC Scopus subject areas

  • Information Systems
  • Management Information Systems
  • Library and Information Sciences
  • Management of Technology and Innovation
  • Strategy and Management

Cite this

Verifying the Proximity and Size Hypothesis for Self-Organizing Maps. / Lin, Chienting; Chen, Hsinchun; Nunamaker, Jay F.

In: Journal of Management Information Systems, Vol. 16, No. 3, 12.1999, p. 57-70.

Research output: Contribution to journalArticle

@article{3b9c4322c683455897fb9d5df004bab5,
title = "Verifying the Proximity and Size Hypothesis for Self-Organizing Maps",
abstract = "The Kohonen Self-Organizing Map (SOM) is an unsupervised learning technique for summarizing high-dimensional data so that similar inputs are, in general, mapped close to one another. When applied to textual data, SOM has been shown to be able to group together related concepts in a data collection and to present major topics within the collection with larger regions. This article presents research in which we sought to validate these properties of SOM, called the Proximity and Size Hypotheses, through a user evaluation study. Building upon our previous research in automatic concept generation and classification, we demonstrated that the Kohonen SOM was able to perform concept clustering effectively, based on its concept precision and recall7 scores as judged by human experts. We also demonstrated a positive relationship between the size of an SOM region and the number of documents contained in the region. We believe this research has established the Kohonen SOM algorithm as an intuitively appealing and promising neural-network-based textual classification technique for addressing part of the longstanding {"}information overload{"} problem.",
keywords = "Document clustering techniques, Experimental research, Group support systems, Self-organizing maps, Unsupervised learning algorithms",
author = "Chienting Lin and Hsinchun Chen and Nunamaker, {Jay F}",
year = "1999",
month = "12",
language = "English (US)",
volume = "16",
pages = "57--70",
journal = "Journal of Management Information Systems",
issn = "0742-1222",
publisher = "M.E. Sharpe Inc.",
number = "3",

}

TY - JOUR

T1 - Verifying the Proximity and Size Hypothesis for Self-Organizing Maps

AU - Lin, Chienting

AU - Chen, Hsinchun

AU - Nunamaker, Jay F

PY - 1999/12

Y1 - 1999/12

N2 - The Kohonen Self-Organizing Map (SOM) is an unsupervised learning technique for summarizing high-dimensional data so that similar inputs are, in general, mapped close to one another. When applied to textual data, SOM has been shown to be able to group together related concepts in a data collection and to present major topics within the collection with larger regions. This article presents research in which we sought to validate these properties of SOM, called the Proximity and Size Hypotheses, through a user evaluation study. Building upon our previous research in automatic concept generation and classification, we demonstrated that the Kohonen SOM was able to perform concept clustering effectively, based on its concept precision and recall7 scores as judged by human experts. We also demonstrated a positive relationship between the size of an SOM region and the number of documents contained in the region. We believe this research has established the Kohonen SOM algorithm as an intuitively appealing and promising neural-network-based textual classification technique for addressing part of the longstanding "information overload" problem.

AB - The Kohonen Self-Organizing Map (SOM) is an unsupervised learning technique for summarizing high-dimensional data so that similar inputs are, in general, mapped close to one another. When applied to textual data, SOM has been shown to be able to group together related concepts in a data collection and to present major topics within the collection with larger regions. This article presents research in which we sought to validate these properties of SOM, called the Proximity and Size Hypotheses, through a user evaluation study. Building upon our previous research in automatic concept generation and classification, we demonstrated that the Kohonen SOM was able to perform concept clustering effectively, based on its concept precision and recall7 scores as judged by human experts. We also demonstrated a positive relationship between the size of an SOM region and the number of documents contained in the region. We believe this research has established the Kohonen SOM algorithm as an intuitively appealing and promising neural-network-based textual classification technique for addressing part of the longstanding "information overload" problem.

KW - Document clustering techniques

KW - Experimental research

KW - Group support systems

KW - Self-organizing maps

KW - Unsupervised learning algorithms

UR - http://www.scopus.com/inward/record.url?scp=0033277412&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0033277412&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:0033277412

VL - 16

SP - 57

EP - 70

JO - Journal of Management Information Systems

JF - Journal of Management Information Systems

SN - 0742-1222

IS - 3

ER -