Interactive Visual Exploration of Local Patterns in Large Scatterplot Spaces

M. Chegini, L. Shao, R. Gregor, D. Lehmann, K. Andrews, T. Schreck

Publikation: Beitrag in einer FachzeitschriftArtikelForschungBegutachtung

Abstract

Analysts often use visualisation techniques like a scatterplot matrix
(SPLOM) to explore multivariate datasets. The scatterplots of a SPLOM
can help to identify and compare two-dimensional global
patterns. However, local patterns which might only exist within
subsets of records are typically much harder to identify and may go
unnoticed among larger sets of plots in a SPLOM. This paper explores
the notion of local patterns and presents a novel approach to visually
select, search for, and compare local patterns in a multivariate
dataset. Model-based and shape-based pattern descriptors are used to
automatically compare local regions in scatterplots to assist in the
discovery of similar local patterns. Mechanisms are provided to
assess the level of similarity between local patterns and to rank
similar patterns effectively. Moreover, a relevance feedback module is
used to suggest potentially relevant local patterns to the user. The
approach has been implemented in an interactive tool and demonstrated
with two real-world datasets and use cases. It supports the discovery
of potentially useful information such as clusters, functional
dependencies between variables, and statistical relationships in
subsets of data records and dimensions.
Originalspracheenglisch
Seiten (von - bis)99-109
Seitenumfang11
FachzeitschriftComputer Graphics Forum
Jahrgang37
Ausgabenummer3
DOIs
PublikationsstatusVeröffentlicht - 2018

Fingerprint

Visualization
Feedback

Fields of Expertise

  • Information, Communication & Computing

Dies zitieren

Interactive Visual Exploration of Local Patterns in Large Scatterplot Spaces. / Chegini, M.; Shao, L.; Gregor, R.; Lehmann, D.; Andrews, K.; Schreck, T.

in: Computer Graphics Forum, Jahrgang 37, Nr. 3, 2018, S. 99-109.

Publikation: Beitrag in einer FachzeitschriftArtikelForschungBegutachtung

@article{cf09d57a21ef460d95ada0c5de46c69b,
title = "Interactive Visual Exploration of Local Patterns in Large Scatterplot Spaces",
abstract = "Analysts often use visualisation techniques like a scatterplot matrix(SPLOM) to explore multivariate datasets. The scatterplots of a SPLOMcan help to identify and compare two-dimensional globalpatterns. However, local patterns which might only exist withinsubsets of records are typically much harder to identify and may gounnoticed among larger sets of plots in a SPLOM. This paper exploresthe notion of local patterns and presents a novel approach to visuallyselect, search for, and compare local patterns in a multivariatedataset. Model-based and shape-based pattern descriptors are used toautomatically compare local regions in scatterplots to assist in thediscovery of similar local patterns. Mechanisms are provided toassess the level of similarity between local patterns and to ranksimilar patterns effectively. Moreover, a relevance feedback module isused to suggest potentially relevant local patterns to the user. Theapproach has been implemented in an interactive tool and demonstratedwith two real-world datasets and use cases. It supports the discoveryof potentially useful information such as clusters, functionaldependencies between variables, and statistical relationships insubsets of data records and dimensions.",
author = "M. Chegini and L. Shao and R. Gregor and D. Lehmann and K. Andrews and T. Schreck",
year = "2018",
doi = "10.1111/cgf.13404",
language = "English",
volume = "37",
pages = "99--109",
journal = "Computer Graphics Forum",
issn = "0167-7055",
publisher = "Wiley-Blackwell",
number = "3",

}

TY - JOUR

T1 - Interactive Visual Exploration of Local Patterns in Large Scatterplot Spaces

AU - Chegini, M.

AU - Shao, L.

AU - Gregor, R.

AU - Lehmann, D.

AU - Andrews, K.

AU - Schreck, T.

PY - 2018

Y1 - 2018

N2 - Analysts often use visualisation techniques like a scatterplot matrix(SPLOM) to explore multivariate datasets. The scatterplots of a SPLOMcan help to identify and compare two-dimensional globalpatterns. However, local patterns which might only exist withinsubsets of records are typically much harder to identify and may gounnoticed among larger sets of plots in a SPLOM. This paper exploresthe notion of local patterns and presents a novel approach to visuallyselect, search for, and compare local patterns in a multivariatedataset. Model-based and shape-based pattern descriptors are used toautomatically compare local regions in scatterplots to assist in thediscovery of similar local patterns. Mechanisms are provided toassess the level of similarity between local patterns and to ranksimilar patterns effectively. Moreover, a relevance feedback module isused to suggest potentially relevant local patterns to the user. Theapproach has been implemented in an interactive tool and demonstratedwith two real-world datasets and use cases. It supports the discoveryof potentially useful information such as clusters, functionaldependencies between variables, and statistical relationships insubsets of data records and dimensions.

AB - Analysts often use visualisation techniques like a scatterplot matrix(SPLOM) to explore multivariate datasets. The scatterplots of a SPLOMcan help to identify and compare two-dimensional globalpatterns. However, local patterns which might only exist withinsubsets of records are typically much harder to identify and may gounnoticed among larger sets of plots in a SPLOM. This paper exploresthe notion of local patterns and presents a novel approach to visuallyselect, search for, and compare local patterns in a multivariatedataset. Model-based and shape-based pattern descriptors are used toautomatically compare local regions in scatterplots to assist in thediscovery of similar local patterns. Mechanisms are provided toassess the level of similarity between local patterns and to ranksimilar patterns effectively. Moreover, a relevance feedback module isused to suggest potentially relevant local patterns to the user. Theapproach has been implemented in an interactive tool and demonstratedwith two real-world datasets and use cases. It supports the discoveryof potentially useful information such as clusters, functionaldependencies between variables, and statistical relationships insubsets of data records and dimensions.

UR - https://ftp.isds.tugraz.at/pub/papers/chegini-eurovis2018-sp-patterns-eg.pdf

U2 - 10.1111/cgf.13404

DO - 10.1111/cgf.13404

M3 - Article

VL - 37

SP - 99

EP - 109

JO - Computer Graphics Forum

JF - Computer Graphics Forum

SN - 0167-7055

IS - 3

ER -