Slide 1

Star CoordinatesKandogan, Star Coordinates

A scatterplot on Star Coordinate system

UNC39Parallel CoordinatesInselberg, Multidimensional detective (parallel coordinates)

UNC40 on steroids

UNC55Multiple Views: Brushing-and-linking

UNC57Effective Table DesignSee Show Me the NumbersProper and effective use of layout, typography, shading, etc. can go a long way(Tables may be underused)UNC5Basic Symbolic DisplaysGraphs ChartsMapsDiagrams

From: S. Kosslyn, Understanding charts and graphs, Applied Cognitive Psychology, 1989.UNC6GraphShowing the relationships between variables values in a data table

UNC7PropertiesGraph Visual display that illustrates one or more relationships among entities Shorthand way to present information Allows a trend, pattern or comparison to be easily comprehended

UNC8IssuesCritical to remain task-centricWhy do you need a graph?What questions are being answered?What data is needed to answer those questions?Who is the audience?

UNC9Graph ComponentsFrameworkMeasurement types, scaleContentMarks, lines, pointsLabelsTitle, axes, ticks

UNC10Many Examples

www.nationmaster.comUNC11Quick Aside Other symbolic displaysChartMapDiagram

UNC12Chart Structure is important, relates entities to each otherPrimarily uses lines, enclosure, position to link entities

Examples: flowchart, family tree, org chart, ...

UNC13MapRepresentation of spatial relationsLocations identified by labels

UNC14DiagramSchematic picture of object or entityParts are symbolic

Examples: figures, steps in a manual, illustrations,...

UNC15Some HistoryWhich is older, map or graph?Maps from about 2300 BCGraphs from 1600sRene DescartesWilliam Playfair, late 1700s

UNC16DetailsWhat are the constituent pieces of these four symbolic displays?What are the building blocks?

UNC17Visual StructuresComposed ofSpatial substrateMarksGraphical properties of marks

UNC18SpaceVisually dominantOften put axes on space to assistUse techniques of composition, alignment, folding, recursion, overloading to 1) increase use of space2) do data encodings

UNC19MarksThings that occur in spacePointsLinesAreasVolumes

UNC20Graphical PropertiesSize, shape, color, orientation...

UNC21Fews Selection & Design ProcessDetermine your message and identify your dataDetermine if a table, or graph, or both is needed to communicate your messageDetermine the best means to encode the valuesDetermine where to display each variableDetermine the best design for the remaining objectsDetermine the range of the quantitative scaleIf a legend is required, determine where to place itDetermine the best location for the quantitative scaleDetermine if grid lines are requiredDetermine what descriptive text is neededDetermine if particular data should be featured and how

S Few Effectively Communicating Numbers http://www.perceptualedge.com/articles/Whitepapers/Communicating_Numbers.pdfUNC22Points, Lines, Bars, BoxesPointsUseful in scatterplots for 2-valuesCan replace bars when scale doesnt start at 0LinesConnect values in a seriesShow changes, trends, patternsNot for a set of nominal or ordinal valuesBarsEmphasizes individual valuesGood for comparing individual valuesBoxesShows a distribution of values

UNC23BarsVertical vs. HorizontalHorizontal can be good if long labels or many items

Multiple BarsCan be used to encode another variable

UNC24Multivariate: Beyond Tables and ChartsData sets of dimensions 1,2,3 are commonNumber of variables per class1 - Univariate data2 - Bivariate data3 - Trivariate data>3 - Hypervariate/Multivariate dataUNC25Univariate DataRepresentations7

5

3

1

Bill020MeanlowhighMiddle 50%Tukey box plotUNC26Bivariate DataRepresentationsScatter plot is commonpricemileageUNC27Trivariate DataRepresentations3D scatter plot is possiblehorsepowermileagepriceUNC28Trivariate3D scatterplot, spin plot2D plot + size (or color)

UNC294D = 3D (spatial) + 1D variable

UNC30So we can do some 4DSpatial 3D plus 1D variable (like tissue density)Spatial 3D plus 1D timeOrthogonal 3D of data (3D plot) plus time

And even 5D (3D spatial, 1D, and 1D time)

Note that many of the 3D spatial ones are best done only if you have 3D capable display. UNC31Different Arrangements of AxesAxes are good Lays out all points in a single spaceposition is 1st in Clevelands rulesUniform treatment of dimensions

Space > 3D ?

Must trash orthogonality

UNC32Multivariate DataNumber of well-known visualization techniques exist for data sets of 1-3 dimensionsline graphs, bar graphs, scatter plots OKWe see a 3-D world (4-D with time)Some visualization for 3,4,5D when some of variables are spatial or time.Interesting (challenging cases) are when we have more variables than this. How best to visualize them?UNC33Map n-D space onto 2-D screenVisual representations:Complex glyphsE.g. star glyphs, faces, embedded visualization, Multiple views of different dimensionsE.g. small multiples, plot matrices, brushing histograms, Spotfire, Non-orthogonal axesE.g. Parallel coords, star coords, Tabular layoutE.g. TableLens, Interactions:Dynamic QueriesBrushing & LinkingSelecting for details, Combinations (combine multiple techniques)UNC34Chernoff FacesEncode different variables values in characteristicsof human facehttp://www.cs.uchicago.edu/~wiseman/chernoff/http://hesketh.com/schampeo/projects/Faces/chernoff.htmlCute applets:

UNC35Star PlotsVar 1Var 2Var 3Var 4Var 5ValueSpace out the nvariables at equalangles around a circle

Each spoke encodesa variables valueUNC37Star Plot examples

http://seamonkey.ed.asu.edu/~behrens/asu/reports/compre/comp1.html

UNC38Parallel Coordinates (2D) Encode variables along a horizontal row Vertical line specifies values

Dataset in a Cartesian graphSame dataset in parallel coordinatesUNC41Parallel Coordinates (4D)Forget about Cartesian orthogonal axes(0,1,-1,2)=0x0y0z0wUNC42Parallel Coordinates Example

BasicGrayscaleColorUNC43

UNCMultiple ViewsGive each variable its own display A B C D E1 4 1 8 3 52 6 3 4 2 13 5 7 2 4 34 2 6 3 1 5A B C D E1234UNC45Small MultiplesNice definitions and examplea from Juice Analytics.UNCSmall Multiples

UNCSmall Multiples

UNCMultiple Graphs--TrellisTrellised visualizations enable you to quickly recognize similarities or differences between different categories in the data. Each individual panel in a trellis visualization displays a subset of the original data table, where the subsets are defined by the categories available in a column or hierarchy.

Two Examples (next slides):Spotfire:For example, if you choose to trellis a visualization based on the two variables "Gender" and "Political affiliation", this will result in four separate panels representing the combinations Female-Republican, Female-Democrat, Male-Republican, and Male-Democrat. If the "Gender" variable is used in conjunction with another variable that has five different values, this will yield ten panels. From this follows that variables with a continuous distribution and a wide range of values (for example, Real values) should be binned before they are used to form a trellis visualization. Otherwise the number of panels quickly becomes unmanageable.

SilverLight: The trellis visualizations allow us to quickly compare data horizontally and vertically with visual sparklines. Not only can you quickly see an individual domain's trend for a region (i.e., domain1 in Europe), but you can also see how domain1.com traffic compares across all three regions. We can also quickly tell if the traffic is meeting our goals by comparing if the trrend line is above or below the KPI line (dotted line).

UNC49

UNC

UNC

UNCSparklinesUse matrix, but in each cell put in not single value, but visual that represents compound element. This way you pack in multiple dimensions into each cell, but can easy scan across cells. Tufte description (originated)MicroSoft Excel examplesInfragistics exampleUNCScatterplot Matrix

Represent each possiblepair of variables in theirown 2-D scatterplot

Useful for what?Misses what?UNC54To Do BetterNeed InteractionSeparate Static from InteractiveVery nice visual index of static presentations is Visualization Zoo

What can we do if we add interaction to the visualizations? In the next section we go further, by adding zoom, filtering, brushing, etc. UNC56Table LensRao, Table Lens

UNC58Table LensSpreadsheet is certainly one hypervariate data presentationIdea: Make the text more visual and symbolicJust leverage basic bar chart ideaUNC59Visual Mapping

Change quantitativevalues to barsUNC60Tricky Part

What do you do fornominal data?UNC61Instantiation

UNC62Details

Focus onitem(s)whileshowingthe contextUNC63See It

http://www.open-video.org/details.php?videoid=8304UNC64FOCUSFeature-Oriented Catalog User InterfaceLeverages spreadsheet metaphor againItems in columns, attributes in rowsUses bars and other representations for attribute valuesUNC65

UNC66CharacteristicsCan sort on any attribute (row)Focus on an attribute value (show only cases having that value) by doubleclicking on itCan type in queries on different attributes to limit what is presented to. Note this is main contribution: dynamic control (selection/change/querying/filtering) of individual attributes. UNC67Limit by Query

UNC68Manifestation

InfoZoomUNC69Categorical data?How about multivariate categorical data?StudentsGender: Female, maleEye color: Brown, blue, green, hazelHair color: Black, red, brown, blonde, grayHome country: USA, China, Italy, India,

UNC70Mosaic Plot

UNC71Mosaic Plot

UNC72Mosaic Plot

UNC73Mosaic PlotReminds you of? (treemaps)

UNC74IBM Attribute ExplorerMultiple histogram views, one per attribute (like trellis)Each data case represented by a squareSquare is positioned relative to that cases value on that attributeSelecting case in one view lights it up in othersQuery sliders for narrowingUse shading to indicate level of query match (darkest for full match)UNC75FeaturesAttribute histogramAll objects on all attribute scalesInteraction with attributes limits

UNC76Features

Inter-relations between attributes brushingUNC77FeaturesColor-encoded sensitivity

UNC78Attribute Explorer

http://www.open-video.org/details.php?videoid=8162UNC79PolarisSee Chris Solte reading for classGood example of integrated control, dynamic filtering, display. Now best seen in Tableau (Chris Solte co-founder with adviser, Pat Hanrahan). UNC80Combining TechniquesMulti-Dimensional + GeoSpatial (DataMaps VT)

UNC811. Small Multiples

1976

Multiple views: 1 attribute / mapUNC822. Embedded Visualizations

Complex glyphs: For each location, show vis of all attributesUNC83Comparison of TechniquesParCood: