Group One

1
CSIS
School of Computer Science and Information Systems
Group One Data Visualization Spring 2005
Doctor of Professional Studies in Computing
2
Agenda
I. Overview II. Foundations of Visualization III.
Visualization and KDD IV. I Can See Clearly
Now V. XmdvTool Demonstration with ISBSG Case
Study
3
I. Overview
Visualize "to form a mental vision, image, or
picture of (something not visible or present to
sight, or of an abstraction) to make visible to
the mind or imagination" The Oxford English
Dictionary, 1989 Many Variations
"Visualization" 1) Visualization in
Scientific Computing (Scientific
Visualization) 2) Information
Visualization 3) Software Visualization
4

• Running Man

5
Fish Eating Boat
6
I. Foundations of Visualization
7
(No Transcript)
8
(No Transcript)
9
(No Transcript)
10
(No Transcript)
11
III. Visualization and KDD

• Knowledge Discovery from Databases
• Data Processing
• Machine Learning
• Evaluation
• Visualization
• Experiments may be nested
• Approach Advocated by YALE
• Yet Another Learning Environment
• http//www-ai.cs.uni-dortmund.de/SOFTWARE/YALE

12
IV. I Can See Clearly Now
Key Points

• Data generation is exploding, particularly
  dimensional data
• Visualization takes place in context tools and
  functionality are driven by user needs and
  objectives
• Yang, et al provide an excellent baseline list of
  core and advanced techniques for consideration
• Keim introduces an interesting 3-dimention view
  linking data type, interaction technique, and
  display type

13
Data Growth Factoids

• How much new information per person? According to
  the Population Reference Bureau, the world
  population is 6.3 billion, thus almost 800 MB of
  recorded information is produced per person each
  year. It would take about 30 feet of books to
  store the equivalent of 800 MB of information on
  paper.
• Information explosion? We estimate that new
  stored information grew about 30 a year between
  1999 and 2002
• The World Wide Web contains about 170 terabytes
  of information on its surface in volume this is
  seventeen times the size of the Library of
  Congress print collections.
• Instant messaging generates five billion messages
  a day (750GB), or 274 Terabytes a year.
• Email generates about 400,000 terabytes of new
  information each year worldwide.

14
Visualization takes place in context different
users with different needs have different
requirement and techniques.
Richness of Information
What should I do ?
Managerial Snap-shot
Interactive reporting
What If" analysis
What Next ?
Use/need
15
Visualization takes place in context different
users with different needs have different
requirement and techniques.
Richness of Information
What should I do ?
Managerial Snap-shot
Interactive reporting
What If" analysis
What Next ?
Use/need
Enterprise Reporting Navigation needs and
reliable information
Mutli-dimensional speed of thought
Prescribed action Alerts and notifications
Managed Metrics Scorecard Dashboards
Analysis and predictive values
Typical Output
16
Visualization takes place in context different
users with different needs have different
requirement and techniques.
Richness of Information
What should I do ?
Managerial Snap-shot
Interactive reporting
What If" analysis
What Next ?
Use/need
Enterprise Reporting Navigation needs and
reliable information
Mutli-dimensional speed of thought
Prescribed action Alerts and notifications
Managed Metrics Scorecard Dashboards
Analysis and predictive values
Typical Output
Filter and Zoom
Slice Dice, Pivot tables
Derived Information
Recommend and Act
Interaction
Fixed display
17
Visualization takes place in context different
users with different needs have different
requirement and techniques.
Richness of Information
What should I do ?
Managerial Snap-shot
Interactive reporting
What If" analysis
What Next ?
Use/need
Enterprise Reporting Navigation needs and
reliable information
Mutli-dimensional speed of thought
Prescribed action Alerts and notifications
Managed Metrics Scorecard Dashboards
Analysis and predictive values
Typical Output
Filter and Zoom
Slice Dice, Pivot tables
Derived Information
Recommend and Act
Interaction
Fixed display
Use the data to prove/disprove a hypothesis
Use the data to generate hypotheses
18
Yang, et al identify Core Navigation Tool

• Filter reduce the amount of data to increase
  focus
• Distortion enlarge some part of a display to
  examine details
• Zooming and Panning enlarge, make smaller, move
  through display
• Manual Pixel re-ordering top to bottom, bottom
  to top
• Comparing create/examine relationships
• Refining generate a new, focused display of
  data subset

19
Yang, et al identify Advance Navigation Tool

• Showing names mouse-overs
• Layer re-ordering ordering of overlapping data
• Manual relocation separation of overlapping
  data
• Extent Scaling interactive, proportional
  resizing
• Dynamic Masking hiding of irrelevant data
• Automatic Shifting automatic overlap reduction

20
Keim creates a 3-dimentional chart that relates
interaction technique, type of data, and
visualization technique
21
Breakdown and examination of Keim model
Simple data
Complex data
22
Breakdown and examination of Keim model
Interaction and manipulation techniques, similar
to Yang
23
Breakdown and examination of Keim model
Recommended display type (some of which we will
see in the demos)
24
V. XmdvTool Demonstration with ISBSG Case Study

• Tool Available at http//davis.wpi.edu/xmdv
• Methods
• Scatterplots
• Glyphs
• Parallel Cordinates
• Dimensional Stacking
• N-D Brush
• Highlight
• Mask
• Values
• Average

25
Source of Case Study

• The International Software Benchmarking Standards
  Group
• Mission Help Improve Management of IT Resources
  Through a Public Repository
• Produces ISBSG Estimating, Benchmarking
  Research Suite (Release 8 in 2003) of Data and
  Tools
• Academic Use Free or Nominal Charge
• Web Site www.isbsg.org
• Same Source As Team Ones Data Mining Project

26
Composition of Study File

• 451 New Development Projects
• Fields
• Size in Adjusted Function Points
• Duration in Months
• Maximum Team Size
• Work Effort in Hours
• Project Delivery Rate