Semi-supervised construction of general visualization hierarchies

Peter Tino, Yi Sun, Ian T. Nabney

Research output: Contribution to conferencePaper

Abstract

We have recently developed a principled approach to interactive non-linear hierarchical visualization [8] based on the Generative Topographic Mapping (GTM). Hierarchical plots are needed when a single visualization plot is not sufficient (e.g. when dealing with large quantities of data). In this paper we extend our system by giving the user a choice of initializing the child plots of the current plot in either interactive, or automatic mode. In the interactive mode the user interactively selects ``regions of interest'' as in [8], whereas in the automatic mode an unsupervised minimum message length (MML)-driven construction of a mixture of GTMs is used. The latter is particularly useful when the plots are covered with dense clusters of highly overlapping data projections, making it difficult to use the interactive mode. Such a situation often arises when visualizing large data sets. We illustrate our approach on a data set of 2300 18-dimensional points and mention extension of our system to accommodate discrete data types.
Original languageEnglish
Pages1-7
Number of pages7
Publication statusPublished - 2002
EventInternational Conference on Artificial Intelligence, 2002 -
Duration: 1 Jan 20021 Jan 2002

Conference

ConferenceInternational Conference on Artificial Intelligence, 2002
Period1/01/021/01/02

Fingerprint

Visualization

Keywords

  • hierarchical visualization
  • Generative Topographic Mapping
  • interactive mode
  • automatic mode
  • overlapping data projections

Cite this

Tino, P., Sun, Y., & Nabney, I. T. (2002). Semi-supervised construction of general visualization hierarchies. 1-7. Paper presented at International Conference on Artificial Intelligence, 2002, .
Tino, Peter ; Sun, Yi ; Nabney, Ian T. / Semi-supervised construction of general visualization hierarchies. Paper presented at International Conference on Artificial Intelligence, 2002, .7 p.
@conference{c2e01a2f343448a9bec716bda4a3d8a3,
title = "Semi-supervised construction of general visualization hierarchies",
abstract = "We have recently developed a principled approach to interactive non-linear hierarchical visualization [8] based on the Generative Topographic Mapping (GTM). Hierarchical plots are needed when a single visualization plot is not sufficient (e.g. when dealing with large quantities of data). In this paper we extend our system by giving the user a choice of initializing the child plots of the current plot in either interactive, or automatic mode. In the interactive mode the user interactively selects ``regions of interest'' as in [8], whereas in the automatic mode an unsupervised minimum message length (MML)-driven construction of a mixture of GTMs is used. The latter is particularly useful when the plots are covered with dense clusters of highly overlapping data projections, making it difficult to use the interactive mode. Such a situation often arises when visualizing large data sets. We illustrate our approach on a data set of 2300 18-dimensional points and mention extension of our system to accommodate discrete data types.",
keywords = "hierarchical visualization, Generative Topographic Mapping, interactive mode, automatic mode, overlapping data projections",
author = "Peter Tino and Yi Sun and Nabney, {Ian T.}",
year = "2002",
language = "English",
pages = "1--7",
note = "International Conference on Artificial Intelligence, 2002 ; Conference date: 01-01-2002 Through 01-01-2002",

}

Tino, P, Sun, Y & Nabney, IT 2002, 'Semi-supervised construction of general visualization hierarchies' Paper presented at International Conference on Artificial Intelligence, 2002, 1/01/02 - 1/01/02, pp. 1-7.

Semi-supervised construction of general visualization hierarchies. / Tino, Peter; Sun, Yi; Nabney, Ian T.

2002. 1-7 Paper presented at International Conference on Artificial Intelligence, 2002, .

Research output: Contribution to conferencePaper

TY - CONF

T1 - Semi-supervised construction of general visualization hierarchies

AU - Tino, Peter

AU - Sun, Yi

AU - Nabney, Ian T.

PY - 2002

Y1 - 2002

N2 - We have recently developed a principled approach to interactive non-linear hierarchical visualization [8] based on the Generative Topographic Mapping (GTM). Hierarchical plots are needed when a single visualization plot is not sufficient (e.g. when dealing with large quantities of data). In this paper we extend our system by giving the user a choice of initializing the child plots of the current plot in either interactive, or automatic mode. In the interactive mode the user interactively selects ``regions of interest'' as in [8], whereas in the automatic mode an unsupervised minimum message length (MML)-driven construction of a mixture of GTMs is used. The latter is particularly useful when the plots are covered with dense clusters of highly overlapping data projections, making it difficult to use the interactive mode. Such a situation often arises when visualizing large data sets. We illustrate our approach on a data set of 2300 18-dimensional points and mention extension of our system to accommodate discrete data types.

AB - We have recently developed a principled approach to interactive non-linear hierarchical visualization [8] based on the Generative Topographic Mapping (GTM). Hierarchical plots are needed when a single visualization plot is not sufficient (e.g. when dealing with large quantities of data). In this paper we extend our system by giving the user a choice of initializing the child plots of the current plot in either interactive, or automatic mode. In the interactive mode the user interactively selects ``regions of interest'' as in [8], whereas in the automatic mode an unsupervised minimum message length (MML)-driven construction of a mixture of GTMs is used. The latter is particularly useful when the plots are covered with dense clusters of highly overlapping data projections, making it difficult to use the interactive mode. Such a situation often arises when visualizing large data sets. We illustrate our approach on a data set of 2300 18-dimensional points and mention extension of our system to accommodate discrete data types.

KW - hierarchical visualization

KW - Generative Topographic Mapping

KW - interactive mode

KW - automatic mode

KW - overlapping data projections

M3 - Paper

SP - 1

EP - 7

ER -

Tino P, Sun Y, Nabney IT. Semi-supervised construction of general visualization hierarchies. 2002. Paper presented at International Conference on Artificial Intelligence, 2002, .