Interactive Visual Analysis of Temporal Cluster Structures

Cagatay Turkay, Julius Parulek, Nathalie Reuter, Helwig Hauser

ARTICLE, Computer Graphics Forum, 2011

• www

Abstract

Cluster analysis is a useful method which reveals underlying structures and relations of items after grouping them into clusters. In the case of temporal data, clusters are defined over time intervals where they usually exhibit structural changes. Conventional cluster analysis does not provide sufficient methods to analyze these structural changes, which are, however, crucial in the interpretation and evaluation of temporal clusters. In this paper, we present two novel and interactive visualization techniques that enable users to explore and interpret the structural changes of temporal clusters. We introduce the temporal cluster view, which visualizes the structural quality of a number of temporal clusters, and temporal signatures, which represents the structure of clusters over time. We discuss how these views are utilized to understand the temporal evolution of clusters. We evaluate the proposed techniques in the cluster analysis of mixed lipid bilayers.

Published

Computer Graphics Forum

• Volume: 30
• Number: 3
• Pages: 711–720
• Event: EuroVis 2011
• Location: Bergen, Norway
• URL: http://dx.doi.org/10.1111/j.1467-8659.2011.01920.x

Media

BibTeX

@article{turkay11temporalCluster,
  author = {Cagatay Turkay and Julius Parulek and Nathalie Reuter and Helwig Hauser},
  title = {Interactive Visual Analysis of Temporal Cluster Structures},
  year = {2011},
abstract = {Cluster analysis is a useful method which reveals underlying 
structures and relations of items after grouping them into clusters. In 
the case of temporal data, clusters are defined over time intervals where 
they usually exhibit structural changes. Conventional cluster analysis 
does not provide sufficient methods to analyze these structural changes, 
which are, however, crucial in the interpretation and evaluation of 
temporal clusters. In this paper, we present two novel and interactive 
visualization techniques that enable users to explore and interpret the 
structural changes of temporal clusters. We introduce the temporal 
cluster view, which visualizes the structural quality of a number of 
temporal clusters, and temporal signatures, which represents the structure 
of clusters over time. We discuss how these views are utilized to 
understand the temporal evolution of clusters. We evaluate the proposed
techniques in the cluster analysis of mixed lipid bilayers.},
  journal = {Computer Graphics Forum},
  volume = {30},
  number = {3},
  pages = {711--720},
  url = {http://dx.doi.org/10.1111/j.1467-8659.2011.01920.x},  
  event = {EuroVis 2011},
  location = {Bergen, Norway},  


}