Making many-to-many parallel coordinate plots scalable by asymmetric biclustering

Hsiang Yun Wu, Yusuke Niibe, Kazuho Watanabe, Shigeo Takahashi, Makoto Uemura, Issei Fujishiro

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

1 Citation (Scopus)

Abstract

Datasets obtained through recently advanced measurement techniques tend to possess a large number of dimensions. This leads to explosively increasing computation costs for analyzing such datasets, thus making formulation and verification of scientific hypotheses very difficult. Therefore, an efficient approach to identifying feature subspaces of target datasets, that is, the subspaces of dimension variables or subsets of the data samples, is required to describe the essence hidden in the original dataset. This paper proposes a visual data mining framework for supporting semiautomatic data analysis that builds upon asymmetric biclustering to explore highly correlated feature subspaces. For this purpose, a variant of parallel coordinate plots, many-to-many parallel coordinate plots, is extended to visually assist appropriate selections of feature subspaces as well as to avoid intrinsic visual clutter. In this framework, biclustering is applied to dimension variables and data samples of the dataset simultaneously and asymmetrically. A set of variable axes are projected to a single composite axis while data samples between two consecutive variable axes are bundled using polygonal strips. This makes the visualization method scalable and enables it to play a key role in the framework. The effectiveness of the proposed framework has been empirically proven, and it is remarkably useful for many-to-many parallel coordinate plots.

Original languageEnglish
Title of host publication2017 IEEE Pacific Visualization Symposium, PacificVis 2017 - Proceedings
PublisherIEEE Computer Society
Pages305-309
Number of pages5
ISBN (Electronic)9781509057382
DOIs
Publication statusPublished - 2017 Sep 11
Event10th IEEE Pacific Visualization Symposium, PacificVis 2017 - Seoul, Korea, Republic of
Duration: 2017 Apr 182017 Apr 21

Other

Other10th IEEE Pacific Visualization Symposium, PacificVis 2017
CountryKorea, Republic of
CitySeoul
Period17/4/1817/4/21

Fingerprint

Data mining
Visualization
Composite materials
Costs

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Computer Vision and Pattern Recognition
  • Hardware and Architecture
  • Software

Cite this

Wu, H. Y., Niibe, Y., Watanabe, K., Takahashi, S., Uemura, M., & Fujishiro, I. (2017). Making many-to-many parallel coordinate plots scalable by asymmetric biclustering. In 2017 IEEE Pacific Visualization Symposium, PacificVis 2017 - Proceedings (pp. 305-309). [8031609] IEEE Computer Society. https://doi.org/10.1109/PACIFICVIS.2017.8031609

Making many-to-many parallel coordinate plots scalable by asymmetric biclustering. / Wu, Hsiang Yun; Niibe, Yusuke; Watanabe, Kazuho; Takahashi, Shigeo; Uemura, Makoto; Fujishiro, Issei.

2017 IEEE Pacific Visualization Symposium, PacificVis 2017 - Proceedings. IEEE Computer Society, 2017. p. 305-309 8031609.

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

Wu, HY, Niibe, Y, Watanabe, K, Takahashi, S, Uemura, M & Fujishiro, I 2017, Making many-to-many parallel coordinate plots scalable by asymmetric biclustering. in 2017 IEEE Pacific Visualization Symposium, PacificVis 2017 - Proceedings., 8031609, IEEE Computer Society, pp. 305-309, 10th IEEE Pacific Visualization Symposium, PacificVis 2017, Seoul, Korea, Republic of, 17/4/18. https://doi.org/10.1109/PACIFICVIS.2017.8031609
Wu HY, Niibe Y, Watanabe K, Takahashi S, Uemura M, Fujishiro I. Making many-to-many parallel coordinate plots scalable by asymmetric biclustering. In 2017 IEEE Pacific Visualization Symposium, PacificVis 2017 - Proceedings. IEEE Computer Society. 2017. p. 305-309. 8031609 https://doi.org/10.1109/PACIFICVIS.2017.8031609
Wu, Hsiang Yun ; Niibe, Yusuke ; Watanabe, Kazuho ; Takahashi, Shigeo ; Uemura, Makoto ; Fujishiro, Issei. / Making many-to-many parallel coordinate plots scalable by asymmetric biclustering. 2017 IEEE Pacific Visualization Symposium, PacificVis 2017 - Proceedings. IEEE Computer Society, 2017. pp. 305-309
@inproceedings{a049fefce44d4ab4a9476837f02154bc,
title = "Making many-to-many parallel coordinate plots scalable by asymmetric biclustering",
abstract = "Datasets obtained through recently advanced measurement techniques tend to possess a large number of dimensions. This leads to explosively increasing computation costs for analyzing such datasets, thus making formulation and verification of scientific hypotheses very difficult. Therefore, an efficient approach to identifying feature subspaces of target datasets, that is, the subspaces of dimension variables or subsets of the data samples, is required to describe the essence hidden in the original dataset. This paper proposes a visual data mining framework for supporting semiautomatic data analysis that builds upon asymmetric biclustering to explore highly correlated feature subspaces. For this purpose, a variant of parallel coordinate plots, many-to-many parallel coordinate plots, is extended to visually assist appropriate selections of feature subspaces as well as to avoid intrinsic visual clutter. In this framework, biclustering is applied to dimension variables and data samples of the dataset simultaneously and asymmetrically. A set of variable axes are projected to a single composite axis while data samples between two consecutive variable axes are bundled using polygonal strips. This makes the visualization method scalable and enables it to play a key role in the framework. The effectiveness of the proposed framework has been empirically proven, and it is remarkably useful for many-to-many parallel coordinate plots.",
author = "Wu, {Hsiang Yun} and Yusuke Niibe and Kazuho Watanabe and Shigeo Takahashi and Makoto Uemura and Issei Fujishiro",
year = "2017",
month = "9",
day = "11",
doi = "10.1109/PACIFICVIS.2017.8031609",
language = "English",
pages = "305--309",
booktitle = "2017 IEEE Pacific Visualization Symposium, PacificVis 2017 - Proceedings",
publisher = "IEEE Computer Society",

}

TY - GEN

T1 - Making many-to-many parallel coordinate plots scalable by asymmetric biclustering

AU - Wu, Hsiang Yun

AU - Niibe, Yusuke

AU - Watanabe, Kazuho

AU - Takahashi, Shigeo

AU - Uemura, Makoto

AU - Fujishiro, Issei

PY - 2017/9/11

Y1 - 2017/9/11

N2 - Datasets obtained through recently advanced measurement techniques tend to possess a large number of dimensions. This leads to explosively increasing computation costs for analyzing such datasets, thus making formulation and verification of scientific hypotheses very difficult. Therefore, an efficient approach to identifying feature subspaces of target datasets, that is, the subspaces of dimension variables or subsets of the data samples, is required to describe the essence hidden in the original dataset. This paper proposes a visual data mining framework for supporting semiautomatic data analysis that builds upon asymmetric biclustering to explore highly correlated feature subspaces. For this purpose, a variant of parallel coordinate plots, many-to-many parallel coordinate plots, is extended to visually assist appropriate selections of feature subspaces as well as to avoid intrinsic visual clutter. In this framework, biclustering is applied to dimension variables and data samples of the dataset simultaneously and asymmetrically. A set of variable axes are projected to a single composite axis while data samples between two consecutive variable axes are bundled using polygonal strips. This makes the visualization method scalable and enables it to play a key role in the framework. The effectiveness of the proposed framework has been empirically proven, and it is remarkably useful for many-to-many parallel coordinate plots.

AB - Datasets obtained through recently advanced measurement techniques tend to possess a large number of dimensions. This leads to explosively increasing computation costs for analyzing such datasets, thus making formulation and verification of scientific hypotheses very difficult. Therefore, an efficient approach to identifying feature subspaces of target datasets, that is, the subspaces of dimension variables or subsets of the data samples, is required to describe the essence hidden in the original dataset. This paper proposes a visual data mining framework for supporting semiautomatic data analysis that builds upon asymmetric biclustering to explore highly correlated feature subspaces. For this purpose, a variant of parallel coordinate plots, many-to-many parallel coordinate plots, is extended to visually assist appropriate selections of feature subspaces as well as to avoid intrinsic visual clutter. In this framework, biclustering is applied to dimension variables and data samples of the dataset simultaneously and asymmetrically. A set of variable axes are projected to a single composite axis while data samples between two consecutive variable axes are bundled using polygonal strips. This makes the visualization method scalable and enables it to play a key role in the framework. The effectiveness of the proposed framework has been empirically proven, and it is remarkably useful for many-to-many parallel coordinate plots.

UR - http://www.scopus.com/inward/record.url?scp=85032025855&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85032025855&partnerID=8YFLogxK

U2 - 10.1109/PACIFICVIS.2017.8031609

DO - 10.1109/PACIFICVIS.2017.8031609

M3 - Conference contribution

SP - 305

EP - 309

BT - 2017 IEEE Pacific Visualization Symposium, PacificVis 2017 - Proceedings

PB - IEEE Computer Society

ER -