Fused visualization of complex information spaces

Publication Type:
Thesis
Issue Date:
2006
Full metadata record
Files in This Item:
Filename Description Size
Thumbnail01Front.pdf6.29 MB
Adobe PDF
Thumbnail02Whole.pdf78.19 MB
Adobe PDF
With the rapid growth of information analysis and data mining technologies, the massive data sets available for access have been merged and refined to manifold information, including raw data and all kinds of analytical results. Since data sets become increasingly complex, the current visual analytical techniques no longer satisfy the needs of exploring and analyzing data. This situation raises the challenges in the current state of information visualization: 1) Due to the complexity of information, sometimes it is unlikely to use a single visual metaphor to model the intricate information well in a single visualization. 2) Each existing visualization method has its own limitations in terms of satisfying domain specific requirements, when dealing with complex data sets. The proposed fused visualization methodology attempts to address the above issues by combining multiple existing visualization techniques in a single visualization. It takes the advantages and reduces the weaknesses of the existing methods. We have successfully applied this methodology to each stage of the proposed Analytical Information Visualization. In particular, three fused visualization techniques are developed to improve the quality of existing techniques. First, a fused visual metaphor that combines two visual metaphors in a single visualization allows users to navigate spatially referenced information across two different metaphors. Second, a fused layout algorithm that combines two graph drawing methods achieves the fast convergence in geometric layout for the force-directed layout algorithm; Third, a fused viewing technique that combines ID and 2D distortional visual viewing methods in one browser resolves the inefficient space utilization problem. Moreover, the fused layout algorithm has been evaluated against other existing force-directed layout algorithms. Two case studies that apply our techniques to an outbreak management system and an online bookstore respectively have been delivered.
Please use this identifier to cite or link to this item: