Interaction design in multidimensional visualization : techniques for multidimensional data visualization, exploration and visual analytics

Huang, T

Interaction design in multidimensional visualization : techniques for multidimensional data visualization, exploration and visual analytics

Huang, T

Permalink

Publication Type:: Thesis
Issue Date:: 2015

Open Access

Copyright Clearance Process

Recently Added
In Progress
Open Access

This item is open access.

Adobe PDF

Download contents and abstractAdobe PDF (231.48 kB)

Adobe PDF

Download thesisAdobe PDF (12.3 MB)

View statistics

Full metadata record

Field	Value	Language
dc.contributor.author	Huang, T
dc.date.accessioned	2015-09-01T03:02:21Z
dc.date.available	2015-09-01T03:02:21Z
dc.date.issued	2015
dc.identifier.uri	http://hdl.handle.net/10453/36969
dc.description	University of Technology Sydney. Faculty of Engineering and Information Technology.	en_US
dc.description.abstract	Interaction is an overloaded term in information visualization. Basically, every software tool is interactive but mostly through the manipulation of a widget. Broadly speaking, a visualization is just a software application. What makes the interactive component of a visualization really distinctive is how well it supports an arbitrary selection of data directly in the interface in order to facilitate subsequent analytic tasks. This is challenging due to over-plotting and visual clutter in the multidimensional space and such phenomenon is commonly known as the curse of dimensionality. Data selection is a frontier of a visualization and too many multidimensional visualizations claiming to be interactive mostly address the change of view without explicitly specifying the core technique of how to materialize such operations. Perhaps, the interactive component is achieved through the traditional widget. To overcome the complexity of truly interacting with multidimensional data for effective visual analytics, we first propose an interactive framework for better understanding of the problem domains. Dynamic data selection is materialized by a novel and sophisticated technique called the Hierarchical Virtual Node which opens an application to interact with data directly in parallel coordinates that would otherwise have been impossible or difficult to achieve by existing methods. It works well even under the circumstance of the curse of dimensionality and offers several advantages over others. For example, the use case only requires a mouse click to select a set of data item(s). To achieve an efficient visual analytics, a set of analytic tasks are also developed in each layer of the proposed framework.	en_US
dc.format	Thesis (PhD)	en_US
dc.language.iso	en	en_US
dc.relation	https://opus.lib.uts.edu.au/bitstream/10453/36969/2/02whole.pdf
dc.rights	info:eu-repo/semantics/openAccess
dc.rights	The author owns the copyright in this thesis including all reproduction and reuse rights for the work. The work may not be altered without the permission of the copyright owner. Attribution is essential when quoting or paraphrasing from this thesis.
dc.rights	au.edu.uts.lib/ppc
dc.title	Interaction design in multidimensional visualization : techniques for multidimensional data visualization, exploration and visual analytics	en_US
dc.type	Thesis
utslib.copyright.status	open_access

Abstract:

Interaction is an overloaded term in information visualization. Basically, every software tool is interactive but mostly through the manipulation of a widget. Broadly speaking, a visualization is just a software application. What makes the interactive component of a visualization really distinctive is how well it supports an arbitrary selection of data directly in the interface in order to facilitate subsequent analytic tasks. This is challenging due to over-plotting and visual clutter in the multidimensional space and such phenomenon is commonly known as the curse of dimensionality. Data selection is a frontier of a visualization and too many multidimensional visualizations claiming to be interactive mostly address the change of view without explicitly specifying the core technique of how to materialize such operations. Perhaps, the interactive component is achieved through the traditional widget. To overcome the complexity of truly interacting with multidimensional data for effective visual analytics, we first propose an interactive framework for better understanding of the problem domains. Dynamic data selection is materialized by a novel and sophisticated technique called the Hierarchical Virtual Node which opens an application to interact with data directly in parallel coordinates that would otherwise have been impossible or difficult to achieve by existing methods. It works well even under the circumstance of the curse of dimensionality and offers several advantages over others. For example, the use case only requires a mouse click to select a set of data item(s). To achieve an efficient visual analytics, a set of analytic tasks are also developed in each layer of the proposed framework.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/36969