Glyph Visualization and Yet Another Tree Visualization

1
Glyph Visualizationand Yet Another Tree
Visualization

• Matt Williams
• InfoVis 533c
• April 3, 2003

2
Overview

• Managing Software with New Visual Representation
• TimeWheel
• 3D Wheel
• InfoBUG
• Summary
• Intermission
• Tree Visualization using SpaceTree
• Guiding Principles
• Results
• Summary

3
Managing Software with New Visual Representation

• Chuah and Eick in Information Visualization
  Proceedings IEEE 1997.

http//www-2.cs.cmu.edu/mei/infobug.htm

• Glyph A symbol, such as a stylized figure or
  arrow on a public sign, that imparts information
  nonverbally.
• http//www.dictionary.com

4
Project Management Issues

• Issues related to large project
• Diversity/Variety
• Correspondence to real world
• Time
• Large Data Volumes

5
Time

• How can we represent time
• Animation?
• Good for easily perceived outliers
• Time Series Graphs?
• Good for comparing across time
• TimeWheel

Mei C. Chuah and Stephan G. Eick. Managing
software with new visual representations. In
Proceedings 97 Information Visualization, pages
30--37. IEEE, 1997. http//citeseer.nj.nec.com/chu
ah97managing.html
6
TimeWheel

• Displays trends for a number of attributes at a
  time
• Visually differentiable result comparing
  increasing vs. decreasing treads
• Maintain Objectedness through Gestalt
  principals

Mei C. Chuah and Stephan G. Eick. Managing
software with new visual representations. In
Proceedings 97 Information Visualization, pages
30--37. IEEE, 1997. http//citeseer.nj.nec.com/chu
ah97managing.html
7
TimeWheel

• Can Take it all in one quick look (preattentive
  pop out??)
• Separable (objectedness)
• Avoids local grouping effects (closure)

Mei C. Chuah and Stephan G. Eick. Managing
software with new visual representations. In
Proceedings 97 Information Visualization, pages
30--37. IEEE, 1997. http//citeseer.nj.nec.com/chu
ah97managing.html
8
Time
Separable?
Mei C. Chuah and Stephan G. Eick. Managing
software with new visual representations. In
Proceedings 97 Information Visualization, pages
30--37. IEEE, 1997. http//citeseer.nj.nec.com/chu
ah97managing.html
9
3D Wheel
Mei C. Chuah and Stephan G. Eick. Managing
software with new visual representations. In
Proceedings 97 Information Visualization, pages
30--37. IEEE, 1997. http//citeseer.nj.nec.com/chu
ah97managing.html
10
InfoBUG

• Show many properties while still maintaining
  objectedness
• Certain patterns preattentively pop out
• Interactive

Mei C. Chuah and Stephan G. Eick. Managing
software with new visual representations. In
Proceedings 97 Information Visualization, pages
30--37. IEEE, 1997. http//citeseer.nj.nec.com/chu
ah97managing.html
11
InfoBUG

• Body
• Bar - file changes
• Spots number of subsystems
• Useful??
• Tail
• Width new lines (errors/functionality)
• Height Lines deleted

• Wings
• Lines of Code vs. Number of Errors
• Interactive
• Head
• Code Type

Mei C. Chuah and Stephan G. Eick. Managing
software with new visual representations. In
Proceedings 97 Information Visualization, pages
30--37. IEEE, 1997. http//citeseer.nj.nec.com/chu
ah97managing.html
12
InfoBUG
Mei C. Chuah and Stephan G. Eick. Managing
software with new visual representations. In
Proceedings 97 Information Visualization, pages
30--37. IEEE, 1997. http//citeseer.nj.nec.com/chu
ah97managing.html
13
Glyph Visualization

• Conclusion
• Using gestalt principals Glyphs can objectify
  properties using and offer trend pattern pop
  out
• Questions??

Mei C. Chuah and Stephan G. Eick. Managing
software with new visual representations. In
Proceedings 97 Information Visualization, pages
30--37. IEEE, 1997. http//citeseer.nj.nec.com/chu
ah97managing.html
14
Intermission

• Hungry?

15
Intermission

• Suffer!

http//www.lardlad.com/cgi-bin/image.cgi?school/sc
hool07.gif
16
Tree Visualization using SpaceTree

• Treemaps, Hyperbolic, and now
• SpaceTree

• SpaceTree Supporting Exploration in Large Node
  Link Tree, Design Evolution and Empirical
  Evaluation Catherine Plaisant, Jesse Grosjean,
  and Ben B. Bederson. Proc. InfoVis 2002.

17
Tree Visualization using SpaceTree

• Navigate by clicking on the nodes
• Aggregation of unopened branches into a triangle
  representing depth and average width. Triangle
  shading represented number of nodes
• Dynamic Filters/Queries

• SpaceTree Supporting Exploration in Large Node
  Link Tree, Design Evolution and Empirical
  Evaluation Catherine Plaisant, Jesse Grosjean,
  and Ben B. Bederson. Proc. InfoVis 2002.

18
SpaceTree - Guiding Principles

• 1) Semantic Zooming vs. Geometric scaling
• Simple progressive scaling provides impractical
  results must be readable
• Instead SpaceTree provides semantic zooming from
  readable nodes to aggregated triangles

• SpaceTree Supporting Exploration in Large Node
  Link Tree, Design Evolution and Empirical
  Evaluation Catherine Plaisant, Jesse Grosjean,
  and Ben B. Bederson. Proc. InfoVis 2002.

19
SpaceTree - Guiding Principles

• 2) Automatic Maximization the number of levels
  shown

• SpaceTree Supporting Exploration in Large Node
  Link Tree, Design Evolution and Empirical
  Evaluation Catherine Plaisant, Jesse Grosjean,
  and Ben B. Bederson. Proc. InfoVis 2002.

20
SpaceTree - Guiding Principles

• 3) Decompose the tree animation Trim, Translate
  and Grow.
• Trim branches that would overlap new branch
• Translate Tree so the new branch will fit
• Grow Tree
• 4) Maintain Landmarks to reduce disorientation
  during animation
• Provide Highlight for the chosen node and path to
  the root

• SpaceTree Supporting Exploration in Large Node
  Link Tree, Design Evolution and Empirical
  Evaluation Catherine Plaisant, Jesse Grosjean,
  and Ben B. Bederson. Proc. InfoVis 2002.

21
SpaceTree - Guiding Principles

• 5) Dynamic filtering/Queries

• SpaceTree Supporting Exploration in Large Node
  Link Tree, Design Evolution and Empirical
  Evaluation Catherine Plaisant, Jesse Grosjean,
  and Ben B. Bederson. Proc. InfoVis 2002.

22
SpaceTree - Guiding Principles

• 6) Data-Aware Zooming Controls
• Allow Zooming basing on node selection rather
  than free zooming

• SpaceTree Supporting Exploration in Large Node
  Link Tree, Design Evolution and Empirical
  Evaluation Catherine Plaisant, Jesse Grosjean,
  and Ben B. Bederson. Proc. InfoVis 2002.

23
SpaceTree - Results

• Experiment comparing Windows Explorer, Hyperbolic
  Browser(Lamping), and SpaceTree.
• Node Finding
• Returning to previously viewed nodes
• Listing ancestors of a node SpaceTree faster
• Find 3 nodes that have more than 10 children
  Hyperbolic faster
• User Preferences coolness factor

• SpaceTree Supporting Exploration in Large Node
  Link Tree, Design Evolution and Empirical
  Evaluation Catherine Plaisant, Jesse Grosjean,
  and Ben B. Bederson. Proc. InfoVis 2002.

24
SpaceTree Summary

• Natural, Clear Navigation of Trees
• Aggregation Semantic Zooming
• Dynamic Queries
• No Breakthrough Results

• SpaceTree Supporting Exploration in Large Node
  Link Tree, Design Evolution and Empirical
  Evaluation Catherine Plaisant, Jesse Grosjean,
  and Ben B. Bederson. Proc. InfoVis 2002.

25
The End
http//www.lardlad.com/cgi-bin/image.cgi?homer/hom
er115.gif