Post on 06-Jul-2020
Visualizing Uncertainty:Computer Science Perspective
Ben Shneiderman, Univ of Maryland, College ParkAlex Pang, Univ of California, Santa Cruz
National Academy of Sciences WorkshopMarch 3-4, 2005, Washington, DC
What do we mean by uncertainty?Why is this an issue now?
March 3-4, 2005NAS Workshop: Visualizing Uncertainty 4
Sources of uncertainty
ÆMeasurement problems ÆRanges/SummariesÆMissing dataÆHuman ratingsÆPotential deceptionsÆPrivacy protectionÆRisk assessmentsÆForecasts
§ Scientific data § Intelligence sources § Statistical analyses§ Medical images§ Gene expression§ Simulations§ Financial models§ Weather § Consumer ratings
March 3-4, 2005NAS Workshop: Visualizing Uncertainty 5
Visualizationsl Text, statistical measures…
l 1D: Lists, documents, numeric ranges…l 2D: Geographic Infol 3D: Scientific Visualization
l Multi-Variate: Information Visualizationl Temporal: Patient histories, web logs …l Tree: Taxonomies, org charts, directories …l Network: Social, communication…
Info
Viz
SciV
iz
March 3-4, 2005NAS Workshop: Visualizing Uncertainty 6
Text: Statistical uncertaintyl Text
l 1Dl 2Dl 3D
l Multi-Vl Temporall Treel Network
* = p<.05 ** = p<.01
*-.124Hours per Week
-.041Computer Years
**.233Time to Fix (Incident)
**.293Time Lost (Incident)
Time Variables
Frustration (N=372)
Margin of error +/- 3%
7%79%Kerry81%14%BushRepsDemsPoll
low1.2 NW
high1.5 SE
med0.8NE
low1.0 SW
ReliabilityRain inches
March 3-4, 2005NAS Workshop: Visualizing Uncertainty 7
Risk/Danger vs. Trust/Validity/Securityl Text
l 1Dl 2Dl 3D
l Multi-Vl Temporall Treel Network
highTechStocks
medBlue ChipStocks
medReal Estate
lowMunicipalBonds
Risk
TerrorThreat levels
Highly About HighlyUnlikely Unlikely Even Likely Likely
March 3-4, 2005NAS Workshop: Visualizing Uncertainty 8
1D: Ranges, variations & forecastsl Text
l 1Dl 2Dl 3D
l Multi-Vl Temporall Treel Network
0
10
20
30
40
50
60
Type 1 Type 2 Type 3
Question Type
Ave
rage
Com
plet
ion
Tim
es
(with
sta
ndar
d de
viat
ions
)
March 3-4, 2005NAS Workshop: Visualizing Uncertainty 9
2D: Ball glyphs Delta (observation - forecast)
http://www.cse.ucsc.edu/research/slvg/assim.html
l Text
l 1Dl 2Dl 3D
l Multi-Vl Temporall Treel Network
March 3-4, 2005NAS Workshop: Visualizing Uncertainty 10
2D: Arrow glyphs (Direction & velocity)
http://www.cse.ucsc.edu/research/avis/unvis.html
March 3-4, 2005NAS Workshop: Visualizing Uncertainty 11
2D: Box glyphs
Schmidt et al., 2004, Underwater Environmental Uncertainty, IEEE CG&Ahttp://csdl.computer.org/comp/mags/cg/2004/05/g5056abs.htm
March 3-4, 2005NAS Workshop: Visualizing Uncertainty 12
2D: Grid with transparency/shading
Lefevre, Pfautz & Joneshttp://ams.confex.com/ams/pdfpapers/82400.pdf
March 3-4, 2005NAS Workshop: Visualizing Uncertainty 13
Off-the-shelf softwarecan give incorrect contourson data with lots of missingvalues.
Modifications to contouring algorithmto account for largenumber of missingvalues.
2D: Isolines with missing values
March 3-4, 2005NAS Workshop: Visualizing Uncertainty 14
2D: Pseudocolor shows mean values
Luo, Kao & Pang, 2003, EuroVishttp://www.soe.ucsc.edu/~pang/op.pdf
March 3-4, 2005NAS Workshop: Visualizing Uncertainty 15
2D: Darkness = uncertainty (high stddev)
Mean = hueSkew = 1/saturationStddev= 1/value
March 3-4, 2005NAS Workshop: Visualizing Uncertainty 16
2D: Separate layer
http://www.cse.ohio-state.edu/~bordoloi/Pubs/pdfCluster.pdf
March 3-4, 2005NAS Workshop: Visualizing Uncertainty 17
2D: Streamlines with binwise +
Luo, Kao & Pang, 2003http://www.soe.ucsc.edu/~pang/op.pdf
March 3-4, 2005NAS Workshop: Visualizing Uncertainty 18
2D: Weather forecast
March 3-4, 2005NAS Workshop: Visualizing Uncertainty 19
2D: NOAA Storm Prediction Center
http://www.spc.noaa.gov/products/
March 3-4, 2005NAS Workshop: Visualizing Uncertainty 20
2D: Dual views & grid lines
(Cedlink & Rhenigas, 2000)(Howard & MacEachren, 1996)Dark shows pollution Dark shows Certainty
March 3-4, 2005NAS Workshop: Visualizing Uncertainty 21
(MacEachren et al., 1998)
Dual maps for Rate and reliability
Bivariate color scheme Double hatch shows unreliable
March 3-4, 2005NAS Workshop: Visualizing Uncertainty 22
http://svs.gsfc.nasa.gov/vis/a000000/a000000/a000010/index.html
Gray shows missing & interpolated value,superior to using black only
Twiddy, R., Cavallo, J., and Shiri, S. 1994. Restorer: A visualization technique for handling missing data. IEEE Visualization 94, 212-216
2D: Gray for missing & interpolation
March 3-4, 2005NAS Workshop: Visualizing Uncertainty 23
Crisp molecular surfaceProbe radius 1.4
Fuzzy molecular surfaceProbe radius 1.4
Crisp molecular surfaceProbe radius 5.0
Fuzzy molecular surfaceProbe radius 5.0
3D: Fuzzy molecular surface: HIV protease
Lee & Varshney (2002), UM Graphics and Visual Informatics Labhttp://www.cs.umd.edu/gvil
l Text
l 1Dl 2Dl 3D
l Multi-Vl Temporall Treel Network
March 3-4, 2005NAS Workshop: Visualizing Uncertainty 24
3D: Fuzzy molecular densities
March 3-4, 2005NAS Workshop: Visualizing Uncertainty 25
3D: Uncertainty ‘dust’
March 3-4, 2005NAS Workshop: Visualizing Uncertainty 26
3D: Color & opacity
March 3-4, 2005NAS Workshop: Visualizing Uncertainty 27
l Text
l 1Dl 2Dl 3D
l Multi-Vl Temporall Treel Network
Multi-V: Database/spreadsheet tables
March 3-4, 2005NAS Workshop: Visualizing Uncertainty 28
l Text
l 1Dl 2Dl 3D
l Multi-Vl Temporall Treel Network
Multi-V: Database/spreadsheet tables
March 3-4, 2005NAS Workshop: Visualizing Uncertainty 29
Temporal: Granularity of time
l Text
l 1Dl 2Dl 3D
l Multi-Vl Temporall Treel Network
March 3-4, 2005NAS Workshop: Visualizing Uncertainty 30
l Text
l 1Dl 2Dl 3D
l Multi-Vl Temporall Treel Network
Temporal: Granularity of time
March 3-4, 2005NAS Workshop: Visualizing Uncertainty 31
l Text
l 1Dl 2Dl 3D
l Multi-Vl Temporall Treel Network
Time
Uncertainty
Hi
Med
Low
Temporal: Granularity of time
March 3-4, 2005NAS Workshop: Visualizing Uncertainty 32
l Text
l 1Dl 2Dl 3D
l Multi-Vl Temporall Treel Network
Temporal: Granularity of time
March 3-4, 2005NAS Workshop: Visualizing Uncertainty 33
Tree: Topology, values & namesl Text
l 1Dl 2Dl 3D
l Multi-Vl Temporall Treel Network
Certainty
Hi
Med
Low
Very Low
- Gene Ontology- Tree of Life- Medical Subject Heading (MeSH)- Chain of command/Org chart
March 3-4, 2005NAS Workshop: Visualizing Uncertainty 34
Tree: Topology, values & namesl Text
l 1Dl 2Dl 3D
l Multi-Vl Temporall Treel Network
Certainty
Hi
Med
Low
Capacity
Hi
Med
Low
March 3-4, 2005NAS Workshop: Visualizing Uncertainty 35
Tree: Topology, values & namesl Text
l 1Dl 2Dl 3D
l Multi-Vl Temporall Treel Network
Certainty
HiMedLow
Leland or Lee
Mike or Michael
Alex or Alan
Barbara
Scott
Ben or Benjamin
Diane or Di
Ed or Edward or Eddie
March 3-4, 2005NAS Workshop: Visualizing Uncertainty 37
Network: Social relationshipl Text
l 1Dl 2Dl 3D
l Multi-Vl Temporall Treel Network
http://prefuse.sourceforge.net/demos-radial.html
March 3-4, 2005NAS Workshop: Visualizing Uncertainty 38 J.A. Brown, McGregor A.J and H-W Braun.
Network: Communication capacity
March 3-4, 2005NAS Workshop: Visualizing Uncertainty 39
l Text
l 1Dl 2Dl 3D
l Multi-Vl Temporall Treel Network
Certainty
Hi
Med
Low
Flow
Hi
Med
Low
Capacity
Hi
Med
Low
Network: Node & edge uncertainty
March 3-4, 2005NAS Workshop: Visualizing Uncertainty 40
Explore novel approaches to:l Text: standard terms, percent, probabilitiesl Box plots, whiskers, rangesl Uncertainty glyphs, isoclines,…l Contours, surfaces, volume renderingl Hue, saturation, value, focus, haze, dust,…l Dual views, probesl Animation, blinking, shaking, flipping,…l Sound, haptics,…
Next steps
March 3-4, 2005NAS Workshop: Visualizing Uncertainty 41
Next stepsl Heighten awareness of the problem among
public, professionals, researchers & developersl Support multi-valued data representation standardsl Explore techniques for each data type l Develop guidelines for implementers:
– Data formats– Interactive interfaces– Visual presentations
l Develop human performance evaluation methods l Publish benchmark datasets & evaluation metricsl Form guidelines for how to
propagate/integrate uncertainty markers