GTM-based data visualisation with incomplete data

Yi Sun, Peter Tino, Ian T. Nabney

Research output: Working paperTechnical report

Abstract

We analyse how the Generative Topographic Mapping (GTM) can be modified to cope with missing values in the training data. Our approach is based on an Expectation -Maximisation (EM) method which estimates the parameters of the mixture components and at the same time deals with the missing values. We incorporate this algorithm into a hierarchical GTM. We verify the method on a toy data set (using a single GTM) and a realistic data set (using a hierarchical GTM). The results show our algorithm can help to construct informative visualisation plots, even when some of the training points are corrupted with missing values.
Original languageEnglish
Place of PublicationBirmingham, UK
PublisherAston University
Number of pages9
ISBN (Print)NCRG/2001/013
Publication statusUnpublished - 2001

Fingerprint

topographic mapping
visualization
method

Keywords

  • Generative Topographic Mapping (GTM)
  • missing values
  • Expectation -Maximisation (EM)
  • hierarchical
  • visualisation plots

Cite this

Sun, Y., Tino, P., & Nabney, I. T. (2001). GTM-based data visualisation with incomplete data. Birmingham, UK: Aston University.
Sun, Yi ; Tino, Peter ; Nabney, Ian T. / GTM-based data visualisation with incomplete data. Birmingham, UK : Aston University, 2001.
@techreport{4c26c40907dd40199cc59e803080e22c,
title = "GTM-based data visualisation with incomplete data",
abstract = "We analyse how the Generative Topographic Mapping (GTM) can be modified to cope with missing values in the training data. Our approach is based on an Expectation -Maximisation (EM) method which estimates the parameters of the mixture components and at the same time deals with the missing values. We incorporate this algorithm into a hierarchical GTM. We verify the method on a toy data set (using a single GTM) and a realistic data set (using a hierarchical GTM). The results show our algorithm can help to construct informative visualisation plots, even when some of the training points are corrupted with missing values.",
keywords = "Generative Topographic Mapping (GTM), missing values, Expectation -Maximisation (EM), hierarchical, visualisation plots",
author = "Yi Sun and Peter Tino and Nabney, {Ian T.}",
year = "2001",
language = "English",
isbn = "NCRG/2001/013",
publisher = "Aston University",
type = "WorkingPaper",
institution = "Aston University",

}

Sun, Y, Tino, P & Nabney, IT 2001 'GTM-based data visualisation with incomplete data' Aston University, Birmingham, UK.

GTM-based data visualisation with incomplete data. / Sun, Yi; Tino, Peter; Nabney, Ian T.

Birmingham, UK : Aston University, 2001.

Research output: Working paperTechnical report

TY - UNPB

T1 - GTM-based data visualisation with incomplete data

AU - Sun, Yi

AU - Tino, Peter

AU - Nabney, Ian T.

PY - 2001

Y1 - 2001

N2 - We analyse how the Generative Topographic Mapping (GTM) can be modified to cope with missing values in the training data. Our approach is based on an Expectation -Maximisation (EM) method which estimates the parameters of the mixture components and at the same time deals with the missing values. We incorporate this algorithm into a hierarchical GTM. We verify the method on a toy data set (using a single GTM) and a realistic data set (using a hierarchical GTM). The results show our algorithm can help to construct informative visualisation plots, even when some of the training points are corrupted with missing values.

AB - We analyse how the Generative Topographic Mapping (GTM) can be modified to cope with missing values in the training data. Our approach is based on an Expectation -Maximisation (EM) method which estimates the parameters of the mixture components and at the same time deals with the missing values. We incorporate this algorithm into a hierarchical GTM. We verify the method on a toy data set (using a single GTM) and a realistic data set (using a hierarchical GTM). The results show our algorithm can help to construct informative visualisation plots, even when some of the training points are corrupted with missing values.

KW - Generative Topographic Mapping (GTM)

KW - missing values

KW - Expectation -Maximisation (EM)

KW - hierarchical

KW - visualisation plots

M3 - Technical report

SN - NCRG/2001/013

BT - GTM-based data visualisation with incomplete data

PB - Aston University

CY - Birmingham, UK

ER -

Sun Y, Tino P, Nabney IT. GTM-based data visualisation with incomplete data. Birmingham, UK: Aston University. 2001.