theory and practice of Data Visualization

1
theory and practice ofData Visualization
Media-X Stanford University
2
Visual AnalysisSoftware
3
Visualization Reference Model
Data
Visual Form
Task
Raw Data
Data Tables
Visual Structures
Views
Data Transformations
Visual Encodings
View Transformations
4
Time-Series Data
5
NameVoyager
http//www.babynamewizard.com/voyager
6
TimeSearcher Hochheiser Shneiderman 02
Based on Wattenbergs 2001 idea for
sketch-based queries of time-series data.
7
Interaction Techniques

• Dynamic Queries
• Filter a visualization through direct, reversable
  actions that avoid complex syntax.

8
Multivariate Data
9
Baseball Statistics from Wills 95
10
Interaction Techniques

• Dynamic Queries
• Filter a visualization through direct, reversable
  actions that avoid complex syntax.
• Brushing and Linking
• Highlight relationships between related items
  across multiple visualization views.

11
GGobi Projections of nD data
http//www.ggobi.org/
12
Dimensionality Reduction
Dimensionality Reduction (Sometimes Considered
Harmful)
13
Centered Projection
14
Parallel Coordinates
15
Parallel Coordinates Inselberg
16
The Multidimensional Detective

• The Dataset
• Production data for 473 batches of a VLSI chip
• 16 process parameters
• X1 The yield of produced chips that are
  useful
• X2 The quality of the produced chips (speed)
• X3 X12 10 types of defects (zero defects
  shown at top)
• X13 X16 4 physical parameters
• The Objective
• Raise the yield (X1) and maintain high quality
  (X2)
• A. Inselberg, Multidimensional Detective,
  Proceedings of IEEE Symposium on Information
  Visualization (InfoVis '97), 1997

17
Parallel Coordinates
18
Inselbergs Principles

• Do not let the picture scare you
• Understand your objectives
• Use them to obtain visual cues
• Carefully scrutinize the picture
• Test your assumptions, especially the I am
  really sure ofs
• You cant be unlucky all the time!

19

• Each line represents a tuple (e.g., VLSI batch)
• Filtered below for high values of X1 and X2

20

• Look for batches with nearly zero defects (9/10)
• Most of these have low yields ? defects OK.

21
(No Transcript)
22

• Notice that X6 behaves differently.
• Allow 2 defects, including X6 ? best batches

23
Parallel Coordinates

• Free implementation Parvis by Ledermen
• http//home.subnet.at/flo/mv/parvis/

24
Tableau / Polaris
25
Polaris

• Research at Stanford by Stolte, Tang, and
  Hanrahan.

26
Tableau
Encodings
Data Display
Data Model
27
Tableau Demo

• The dataset
• Federal Elections Commission Receipts
• Every Congressional Candidate from 1996 to 2002
• 4 Election Cycles
• 9216 Candidacies

28
Hypotheses?

• What might we learn from this data?
• ??

29
Hypotheses?

• What might we learn from this data?
• Correlation between receipts and winners?
• Do receipts increase over time?
• Which states spend the most?
• Which party spends the most?
• Margin of victory vs. amount spent?
• Amount spent between competitors?

30
Tableau Demo
31
Polaris/Tableau Approach

• Insight can simultaneously specify both database
  queries and visualization
• (c.f., Leland Wilkinsons Grammar of Graphics)
• Choose data, then visualization, not vice versa
• Use smart defaults for visual encodings
• More recently automate visualization design

32
Ordinal - Ordinal

• Ordinal - Ordinal

33
Quantitative - Quantitative
34
Ordinal - Quantitative
35
Querying the Database
36
All Marital Status
2000
Year
1990
1980
1970
60
40-59
Age
20-39
Sum along Marital Status
0-19
Single
Married
Divorced
Widowed
Sum along Age
Marital Status
All Ages
All Years
Sum along Year
37
All Marital Status
2000
Year
1990
Roll-Up
1980
1970
60
Drill-Down
40-59
Age
20-39
Sum along Marital Status
0-19
Single
Married
Divorced
Widowed
Sum along Age
Marital Status
All Ages
All Years
Sum along Year
38
(No Transcript)
39
Visual AnalysisExercise
40
Visual Analysis Exercise

• Goal Gain familiarity with one or more visual
  analysis tools to explore data of interest.
• Pick one or more data sets to investigate
• Formulate initial questions or hypotheses
• Select and use visual tools to explore data
• Record findings and experiences, note how
  questions are refined or generated
• Iterate!

41
Visual Analysis Exercise

• Choose one question of interest and prepare a
  final visualization to communicate your findings.
  
• After you have constructed the visualization,
  write a caption and short paragraph describing
  the visualization and what it reveals.
• Think of the figure, the caption and the text as
  material you might include in a research article.

42
Data Set Possibilities

• Use your own data
• Any one interested in sharing?

43
Data Set Possibilities

• Use your own data
• Explore one of our sample data sets
• hci.stanford.edu/jheer/workshop/data
• Census database
• Crime statistics
• Election tallies and contributions

44
Data Set Possibilities

• Use your own data
• Explore one of our sample data sets
• Download public data from the web

45
Tools
46
Trial Version of Tableau

• tableausoftware.com/visualization-workshop
• Tableau supports data in Excel spreadsheets, CSV
  files, or existing relational databases

47
Visualization Tools

• Many-Eyes http//many-eyes.com
• Verfiable http//verifiable.com
• GGobi http//ggobi.org
• Parvis http//home.subnet.at/flo/mv/parvis
• TimeSearcher http//www.cs.umd.edu/hcil/timesearch
  er
• Improvise http//www.cs.ou.edu/weaver/improvise/
  
• GGPlot2 (in R) http//had.co.nz/ggplot2/
• and many others

48
Tree / Network Tools

• GraphViz http//www.graphviz.org
• NodeXL http//www.codeplex.com/NodeXL
• GUESS http//graphexploration.cond.org/
• Pajek http//pajek.imfm.si/doku.php
• TreeMap http//www.cs.umd.edu/hcil/treemap
• Workbench http//nwb.slis.indiana.edu/