Multivariate Visualization Tools
CS 7450 - Information Visualization January 20, 2004
John Stasko
HW 2
• What did you think?
− Difficult?
− Challenges?
• Notion of a particular visualization for this problem (design contribution) versus a more general visualization that could apply to other areas (research
contribution)
Spring 2004 CS 7450 3
Last Time
• We examined a number of
techniques/visualizations for presenting multivariate (typically >3) data sets
− For < 3 variables, the tried and true charting techniques are hard to beat
Techniques
Scatterplot Matrix Chernoff Faces Star Plots
Spring 2004 CS 7450 5
Today
• One more interesting technique
• Systems designed for hypervariate data sets
Another Technique
• Database of data items, each of n dimensions
• Issue a query that specifies a target value of the dimensions
• Often get back no exact matches
• Want to find near matches
Taken from:
D. Keim, H-P Kriegel, “VisDB Database Exploration Using Multid Vis”, IEEE CG&A, 1994.
Spring 2004 CS 7450 7
Relevance Factor
• How close an item is to the query
• Data items have some value that can be numerically quantified
• Each dimension is some distance away from query item
• Sum these up for total distance
• Relevance is inverse of distance
Example
• 5 dimensions, integers 0->255
• Query: 6, 210, 73, 45, 92
• Data item: 8, 200, 73, 50, 91
• Distance: 2 + 10 + 0 + 5 + 1 = 18
• Relevance: 1275 - 18 = 1267
Spring 2004 CS 7450 9
Issues
• What if dimensions are real numbers or text strings?
• What if they’re the same type, but of different orders of magnitude?
• Have to define some kind of distance, then a weight function to multiply by
Technique
• Calculate relevance of all data points
• Sort items based on relevance
• Use spiral technique to order the values
• Color items based on relevance
Spring 2004 CS 7450 11
Relevance Colors
High Low
Empirically established
Spiral Method
Highest relevance value in center, decreasing values grow outward
Spring 2004 CS 7450 13
Display Methodology
Example: five-dimensional data
Total
relevance Dim 1 Dim 2
Dim 3 Dim 4
Dim 5
Spiral in each window
Items ordered by total relevance
Same item appears in same place in each window
Figure from Paper
Spring 2004 CS 7450 15
Example Display
Alternative
• Grouping arrangement
• Doesn’t use multiple windows
• Create all relevance dimensional
depictions for an item and group them
• Spiral out the different data items’
depictions
Spring 2004 CS 7450 17
Grouping Arrangement
Example Display
Multi-window Grouping
8 dimensions 1000 items
Spring 2004 CS 7450 19
Tools
• Examine systems that have been built to visualize (relatively) high-dimensional data sets
− Use some of these existing techniques and add new ones
− Interactivity is a key
• Demos of each of the systems
Spotfire
Spring 2004 CS 7450 21
SeeIt
Eureka
Spring 2004 CS 7450 23
InfoZoom
EZChooser
Spring 2004 CS 7450 25
Grokker
HW 3
• Work with, understand, and critique these systems
− Data sets
− Consider user tasks
Spring 2004 CS 7450 27
HW 1
• Turn back
• Trends I noticed
Upcoming
• Visual perception
− Reading:
Chapter 4
• Tufte’s principles
− Be reading his book
