Feed-forward neural networks and topographic mappings for exploratory data analysis

David Lowe, Michael E Tipping

Research output: Contribution to journalArticle

Abstract

A recent novel approach to the visualisation and analysis of datasets, and one which is particularly applicable to those of a high dimension, is discussed in the context of real applications. A feed-forward neural network is utilised to effect a topographic, structure-preserving, dimension-reducing transformation of the data, with an additional facility to incorporate different degrees of associated subjective information. The properties of this transformation are illustrated on synthetic and real datasets, including the 1992 UK Research Assessment Exercise for funding in higher education. The method is compared and contrasted to established techniques for feature extraction, and related to topographic mappings, the Sammon projection and the statistical field of multidimensional scaling.
Original languageEnglish
Pages (from-to)83-95
Number of pages13
JournalNeural Computing and Applications
Volume4
Issue number2
DOIs
Publication statusPublished - Jun 1996

Fingerprint

Feedforward neural networks
Feature extraction
Visualization
Education

Bibliographical note

The original publication is available at www.springerlink.com

Keywords

  • neural networks
  • topographic mappings
  • data analysis
  • feature extraction
  • sammon mapping
  • multidimensional scaling

Cite this

@article{96c0aaa15d7342f48d17aada80af5172,
title = "Feed-forward neural networks and topographic mappings for exploratory data analysis",
abstract = "A recent novel approach to the visualisation and analysis of datasets, and one which is particularly applicable to those of a high dimension, is discussed in the context of real applications. A feed-forward neural network is utilised to effect a topographic, structure-preserving, dimension-reducing transformation of the data, with an additional facility to incorporate different degrees of associated subjective information. The properties of this transformation are illustrated on synthetic and real datasets, including the 1992 UK Research Assessment Exercise for funding in higher education. The method is compared and contrasted to established techniques for feature extraction, and related to topographic mappings, the Sammon projection and the statistical field of multidimensional scaling.",
keywords = "neural networks, topographic mappings, data analysis, feature extraction, sammon mapping, multidimensional scaling",
author = "David Lowe and Tipping, {Michael E}",
note = "The original publication is available at www.springerlink.com",
year = "1996",
month = "6",
doi = "10.1007/BF01413744",
language = "English",
volume = "4",
pages = "83--95",
journal = "Neural Computing and Applications",
issn = "0941-0643",
publisher = "Springer",
number = "2",

}

Feed-forward neural networks and topographic mappings for exploratory data analysis. / Lowe, David; Tipping, Michael E.

In: Neural Computing and Applications, Vol. 4, No. 2, 06.1996, p. 83-95.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Feed-forward neural networks and topographic mappings for exploratory data analysis

AU - Lowe, David

AU - Tipping, Michael E

N1 - The original publication is available at www.springerlink.com

PY - 1996/6

Y1 - 1996/6

N2 - A recent novel approach to the visualisation and analysis of datasets, and one which is particularly applicable to those of a high dimension, is discussed in the context of real applications. A feed-forward neural network is utilised to effect a topographic, structure-preserving, dimension-reducing transformation of the data, with an additional facility to incorporate different degrees of associated subjective information. The properties of this transformation are illustrated on synthetic and real datasets, including the 1992 UK Research Assessment Exercise for funding in higher education. The method is compared and contrasted to established techniques for feature extraction, and related to topographic mappings, the Sammon projection and the statistical field of multidimensional scaling.

AB - A recent novel approach to the visualisation and analysis of datasets, and one which is particularly applicable to those of a high dimension, is discussed in the context of real applications. A feed-forward neural network is utilised to effect a topographic, structure-preserving, dimension-reducing transformation of the data, with an additional facility to incorporate different degrees of associated subjective information. The properties of this transformation are illustrated on synthetic and real datasets, including the 1992 UK Research Assessment Exercise for funding in higher education. The method is compared and contrasted to established techniques for feature extraction, and related to topographic mappings, the Sammon projection and the statistical field of multidimensional scaling.

KW - neural networks

KW - topographic mappings

KW - data analysis

KW - feature extraction

KW - sammon mapping

KW - multidimensional scaling

UR - http://www.springerlink.com/content/v613273154544272/?p=3bb253f2645a4dffb68eee7b3b1debae&pi=3

U2 - 10.1007/BF01413744

DO - 10.1007/BF01413744

M3 - Article

VL - 4

SP - 83

EP - 95

JO - Neural Computing and Applications

JF - Neural Computing and Applications

SN - 0941-0643

IS - 2

ER -