Abstract visualization of large-scale time-varying data
1 online resource (200 pages) : PDF
University of North Carolina at Charlotte
The explosion of large-scale time-varying datasets has created critical challenges for scientists to study and digest. One core problem for visualization is to develop effective approaches that can be used to study various data features and temporal relationships among large-scale time-varying datasets.In this dissertation, we first present two abstract visualization approaches to visualizing and analyzing time-varying datasets. The first approach visualizes time-varying datasets with succinct lines to represent temporal relationships of the datasets. A time line visualizes time steps as points and temporal sequence as a line. They are generated by sampling the distributions of virtual words across time to study temporal features. The key idea of time line is to encode various data properties with virtual words. We apply virtual words to characterize feature points and use their distribution statistics to measure temporal relationships. The second approach is ensemble visualization, which provides a highly abstract platform for visualizing an ensemble of datasets. Both approaches can be used for exploration, analysis, and demonstration purposes.The second component of this dissertation is an animated visualization approach to study dramatic temporal changes. Animation has been widely used to show trends, dynamic features and transitions in scientific simulations, while animated visualization is new. We present an automatic animation generation approach that simulates the composition and transition of storytelling techniques and synthesizes animations to describe various event features. We also extend the concept of animated visualization to non-traditional time-varying datasets - network protocols - for visualizing key information in abstract sequences. We have evaluated the effectiveness of our animated visualization with a formal user study and demonstrated the advantages of animated visualization for studying time-varying datasets.
Ribarsky, WilliamFan, JianpingSubramanian, KalpathiChen, WeiSun, Yanqing
Thesis (Ph.D.)--University of North Carolina at Charlotte, 2012.
This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s). For additional information, see http://rightsstatements.org/page/InC/1.0/.
Copyright is held by the author unless otherwise indicated.