Multi Dimensional Data Visualization 2 (PowerPoint)

							1D & 2D Spaces for
 Representing Data

     Mao Lin Huang
1-D Representation of Data
1-D Textual Data
               Keyhole Problem
• No context
• Lost, disoriented

•   Where am I?
•   Where can I go?
•   Where do I want to go?
•   How do I get there?
                   Visual Overview
•   Map, organization (spatial layout of concepts)
•   What information is (not) available?
•   Adds context info, relationships
•   Enables direct access
•   Encourages exploration
•   HCI metrics:
        • Improves user performance, learning time, error rates,
          retention, satisfaction
          Navigation Approaches
• Detail Only


• Zooming


• Overview+Detail


• Focus+Context (Distortion, fisheye)
       1D Visual Representation
• Plaisant, “Lifelines”, pp 285
       See personal history




• Mackinlay, “Perspective Wall”, web
     1D Visual Representation
• Eick, “SeeSoft”, p 419
       » Analyze 50,000 lines of code simultaneously by mapping each
         line of code into a thin row.




• Eick, “Data Visualization Sliders”, p 251 (2 pages)
            Navigation Strategies
• Detail Only


• Zooming


• Overview+Detail


• Focus+Context (Distortion, fisheye)
                             Fisheye Menus
• http://www.cs.umd.edu/hcil/fisheyemenu/

• Very Fast
        • due to mouse mechanics, no clicking, mostly vertical sliding
• Alphabet overview helpful
• Fisheye context not useful in this case?
        • Might be more useful in SeeSoft where miniature
          representation gives important information
        • Limits # of readable items to ~10
        • Wasted space at top- and bottom- right
        • Distortion problematic?
        • Alphabet overview distorted at A and Z
        • Scale limited?
• Possible improvement:
        • Same alphabet overview (without end distortion)
        • Remove fisheye, maximize readable items like scrolling
          version
        • Same fast mouse mechanics, scroll fast on left, no scroll on
          right
       Music Animation Machine




• Good for visualizing music during serial playback,
  relate audio to visual structure
• Visualizing entire compositions: increase
  information density
     • See patterns of phrases, instruments, etc.
2-D Representation of Data
                   2-D
• Image browsing
• Maps
                        Today
• Bederson, “Pad++”, p 530
      » a zooming graphic interface to replace icon-based window
        interface




• Furnas, “Space-Scale Diagrams”, web
Space-Scale
               Pad++ on edge
• Like ray-tracing          Info surface



                     zoom




                            window
            Semantic Zooming
• Zooming in, red object turns to blue
              Multiple Views
• Zoom factor ~ 20
      Multiple levels = large scale
• Zoom factor = 20 * 20 * 20 = 8000
Multiple Foci
            Multiple Overviews
• Can have different information types at each level
            2-D + Attributes
• Dynamaps: dynamic queries on maps
 2-D: Focus+Context
Representation of Data
                      2-D
• Robertson, “Document Lens”, p 562
       »




• Spence, “Bifocal Lens”, p 331,333
               Focus+Context
• Details within overview

• “Distortion-oriented display”
• “Fisheye”

• Leung, Apperley, “Taxonomy of distortion-
  oriented presentations”, book pg 350
     Visual Transfer Functions


                       Display
                       surface




                                 Information surface



Identity function =                  Bifocal
normal flat overview
Magnification Functions




      1st Derivative
             Bifocal Display
• Spence, Apperley
                 Bifocal Display
Disadvantage: 1 dimensional
stretching on the 4 sides
Perspective Wall / Document Lens
                 NonLinear Magnification
•   http://www.cs.indiana.edu/hyplan/tkeahey/research/nlm/nlm.html
•   http://www.cs.indiana.edu/hyplan/tkeahey/research/papers/infovis.98.html
                       “Bubble”
Disadvantage: local context highly de-magnified
       “Fisheye”, “wide-angle lens”
Disadvantage: no flat area
             Quiz: TableLens




• Bifocal!
              Fisheye Menus
• Non-linear:
  combination of
  Bubble + fisheye
       Why not magnifying glass?
• Hides local context
             F+C                       vs.              O+D
                                        • + Scales up to larger data (zoom factor
• + Space efficient                     and chaining)
• + Detail connected to context         • + Multi foci easier
       • Smooth transition              • + multiple overviews possible
• + matches human vision/processing?
                                        • + Easy to implement, Less math!
                                             •Fast system performance
• - Distortion                          • - >=2 places to look (cross-eyed!)
• - Longer learning time                     •Tracking field-of-view box hard
• - no flat overview - Need a way to
                                             •Hand-eye coordination problem
  turn off focus
• - Content moves differently than      • - detail and overview disconnected
  mouse                                 • - Windows/space management
• - hard to tell zoom factor
                                        • - replicates detail data in overview