Spectral-based contractible parallel coordinates

Koto Nohno, Hsiang Yun Wu, Kazuho Watanabe, Shigeo Takahashi, Issei Fujishiro

Research output: Chapter in Book/Report/Conference proceedingConference contribution

9 Citations (Scopus)

Abstract

Parallel coordinates is well-known as a popular tool for visualizing the underlying relationships among variables in high-dimension datasets. However, this representation still suffers from visual clutter arising from intersections among poly line plots especially when the number of data samples and their associated dimension become high. This paper presents a method of alleviating such visual clutter by contracting multiple axes through the analysis of correlation between every pair of variables. In this method, we first construct a graph by connecting axis nodes with an edge weighted by data correlation between the corresponding pair of dimensions, and then reorder the multiple axes by projecting the nodes onto the primary axis obtained through the spectral graph analysis. This allows us to compose a dendrogram tree by recursively merging a pair of the closest axes one by one. Our visualization platform helps the visual interpretation of such axis contraction by plotting the principal component of each data sample along the composite axis. Smooth animation of the associated axis contraction and expansion has also been implemented to enhance the visual readability of behavior inherent in the given high-dimensional datasets.

Original languageEnglish
Title of host publicationProceedings of the International Conference on Information Visualisation
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages7-12
Number of pages6
ISBN (Print)9781479941032
DOIs
Publication statusPublished - 2014 Sep 18
Event2014 18th International Conference on Information Visualisation: Visualisation, BioMedical Visualization, Visualisation on Built and Rural Environments and Geometric Modelling and Imaging, IV 2014 - Paris, France
Duration: 2014 Jul 162014 Jul 18

Other

Other2014 18th International Conference on Information Visualisation: Visualisation, BioMedical Visualization, Visualisation on Built and Rural Environments and Geometric Modelling and Imaging, IV 2014
CountryFrance
CityParis
Period14/7/1614/7/18

Fingerprint

Animation
Merging
Visualization
Composite materials

Keywords

  • axis contraction
  • dendrograms
  • parallel coordinates
  • spectral graph theory

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Software
  • Signal Processing

Cite this

Nohno, K., Wu, H. Y., Watanabe, K., Takahashi, S., & Fujishiro, I. (2014). Spectral-based contractible parallel coordinates. In Proceedings of the International Conference on Information Visualisation (pp. 7-12). [6902872] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IV.2014.60

Spectral-based contractible parallel coordinates. / Nohno, Koto; Wu, Hsiang Yun; Watanabe, Kazuho; Takahashi, Shigeo; Fujishiro, Issei.

Proceedings of the International Conference on Information Visualisation. Institute of Electrical and Electronics Engineers Inc., 2014. p. 7-12 6902872.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Nohno, K, Wu, HY, Watanabe, K, Takahashi, S & Fujishiro, I 2014, Spectral-based contractible parallel coordinates. in Proceedings of the International Conference on Information Visualisation., 6902872, Institute of Electrical and Electronics Engineers Inc., pp. 7-12, 2014 18th International Conference on Information Visualisation: Visualisation, BioMedical Visualization, Visualisation on Built and Rural Environments and Geometric Modelling and Imaging, IV 2014, Paris, France, 14/7/16. https://doi.org/10.1109/IV.2014.60
Nohno K, Wu HY, Watanabe K, Takahashi S, Fujishiro I. Spectral-based contractible parallel coordinates. In Proceedings of the International Conference on Information Visualisation. Institute of Electrical and Electronics Engineers Inc. 2014. p. 7-12. 6902872 https://doi.org/10.1109/IV.2014.60
Nohno, Koto ; Wu, Hsiang Yun ; Watanabe, Kazuho ; Takahashi, Shigeo ; Fujishiro, Issei. / Spectral-based contractible parallel coordinates. Proceedings of the International Conference on Information Visualisation. Institute of Electrical and Electronics Engineers Inc., 2014. pp. 7-12
@inproceedings{acaa1c5d2dcc47aab665e6fa2e1078ea,
title = "Spectral-based contractible parallel coordinates",
abstract = "Parallel coordinates is well-known as a popular tool for visualizing the underlying relationships among variables in high-dimension datasets. However, this representation still suffers from visual clutter arising from intersections among poly line plots especially when the number of data samples and their associated dimension become high. This paper presents a method of alleviating such visual clutter by contracting multiple axes through the analysis of correlation between every pair of variables. In this method, we first construct a graph by connecting axis nodes with an edge weighted by data correlation between the corresponding pair of dimensions, and then reorder the multiple axes by projecting the nodes onto the primary axis obtained through the spectral graph analysis. This allows us to compose a dendrogram tree by recursively merging a pair of the closest axes one by one. Our visualization platform helps the visual interpretation of such axis contraction by plotting the principal component of each data sample along the composite axis. Smooth animation of the associated axis contraction and expansion has also been implemented to enhance the visual readability of behavior inherent in the given high-dimensional datasets.",
keywords = "axis contraction, dendrograms, parallel coordinates, spectral graph theory",
author = "Koto Nohno and Wu, {Hsiang Yun} and Kazuho Watanabe and Shigeo Takahashi and Issei Fujishiro",
year = "2014",
month = "9",
day = "18",
doi = "10.1109/IV.2014.60",
language = "English",
isbn = "9781479941032",
pages = "7--12",
booktitle = "Proceedings of the International Conference on Information Visualisation",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Spectral-based contractible parallel coordinates

AU - Nohno, Koto

AU - Wu, Hsiang Yun

AU - Watanabe, Kazuho

AU - Takahashi, Shigeo

AU - Fujishiro, Issei

PY - 2014/9/18

Y1 - 2014/9/18

N2 - Parallel coordinates is well-known as a popular tool for visualizing the underlying relationships among variables in high-dimension datasets. However, this representation still suffers from visual clutter arising from intersections among poly line plots especially when the number of data samples and their associated dimension become high. This paper presents a method of alleviating such visual clutter by contracting multiple axes through the analysis of correlation between every pair of variables. In this method, we first construct a graph by connecting axis nodes with an edge weighted by data correlation between the corresponding pair of dimensions, and then reorder the multiple axes by projecting the nodes onto the primary axis obtained through the spectral graph analysis. This allows us to compose a dendrogram tree by recursively merging a pair of the closest axes one by one. Our visualization platform helps the visual interpretation of such axis contraction by plotting the principal component of each data sample along the composite axis. Smooth animation of the associated axis contraction and expansion has also been implemented to enhance the visual readability of behavior inherent in the given high-dimensional datasets.

AB - Parallel coordinates is well-known as a popular tool for visualizing the underlying relationships among variables in high-dimension datasets. However, this representation still suffers from visual clutter arising from intersections among poly line plots especially when the number of data samples and their associated dimension become high. This paper presents a method of alleviating such visual clutter by contracting multiple axes through the analysis of correlation between every pair of variables. In this method, we first construct a graph by connecting axis nodes with an edge weighted by data correlation between the corresponding pair of dimensions, and then reorder the multiple axes by projecting the nodes onto the primary axis obtained through the spectral graph analysis. This allows us to compose a dendrogram tree by recursively merging a pair of the closest axes one by one. Our visualization platform helps the visual interpretation of such axis contraction by plotting the principal component of each data sample along the composite axis. Smooth animation of the associated axis contraction and expansion has also been implemented to enhance the visual readability of behavior inherent in the given high-dimensional datasets.

KW - axis contraction

KW - dendrograms

KW - parallel coordinates

KW - spectral graph theory

UR - http://www.scopus.com/inward/record.url?scp=84912075377&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84912075377&partnerID=8YFLogxK

U2 - 10.1109/IV.2014.60

DO - 10.1109/IV.2014.60

M3 - Conference contribution

SN - 9781479941032

SP - 7

EP - 12

BT - Proceedings of the International Conference on Information Visualisation

PB - Institute of Electrical and Electronics Engineers Inc.

ER -