hypervariate data visualization · • provide (initial ideas of) data source and format of the...

Post on 15-Sep-2020

1 Views

Category:

Documents

0 Downloads

Preview:

Click to see full reader

TRANSCRIPT

INFOVIS8803DV > SPRING 18

HYPERVARIATE DATA VISUALIZATION

Prof. Rahul C. BasoleCS/MGT 8803-DV > February 5, 2018

INFOVIS8803DV > SPRING 18

Hypervariate Data (n > 3)

• Many well-known visualization techniques exist for datasets of 1-3 dimensions. We have discussed them throughout the semester.

• But what about data sets with more than 3 variables (n>3)?

– These are often the interesting & challenging ones– Many problems are of low-dimensional hypervariate nature (4-20)– How would one visualize such data?

INFOVIS8803DV > SPRING 18

Consider the following Design Challenge

• Data set of 500 cases

• Attributes– 5 quantitative– 4 nominal– 2 ordinal

• Design a visual representation

INFOVIS8803DV > SPRING 18

Here is one crazy idea …

• Use a Spreadsheet– Each variable is positioned into a column– Data cases are rows– This is a projection (mapping)

INFOVIS8803DV > SPRING 18

… or we can use Multiple Views

A B C D E

Case 1

Case 2

Case 3

Case 4

Give each variable its own display.

A B C D E

1 4 1 8 3 5

2 6 3 4 2 1

3 5 7 2 4 3

4 2 6 3 1 5

INFOVIS8803DV > SPRING 18

Scatterplot Matrix (n>3)

Represent each possible pair of variables in their own 2D scatterplot. Good if pairwise correlation is key.

INFOVIS8803DV > SPRING 18

But what if we want to handle all datasets generically?

INFOVIS8803DV > SPRING 18

Recall: Use Iconic Representations

• Glyph (graphical object) represents a data case • Visual properties of glyph represent different variables

INFOVIS8803DV > SPRING 18

INFOVIS8803DV > SPRING 18

Chernoff Face (n>3)

Encode different variables’ values in characteristics of human face

n=11

Herman Chernoff (1973). "The Use of Faces to Represent Points in K-Dimensional Space Graphically". Journal of the American Statistical Association, 68 (342): 361–368.

INFOVIS8803DV > SPRING 18

http://flowingdata.com/2010/08/31/how-to-visualize-data-with-cartoonish-faces/crime-chernoff-faces-by-state-edited-2/

INFOVIS8803DV > SPRING 18

Star Plots (n>3)

INFOVIS8803DV > SPRING 18

Small Multiples of Star Plots

https://andrewpwheeler.files.wordpress.com/2012/03/infographic_star_plots.png

INFOVIS8803DV > SPRING 18

Glyphs (n > 3)

• Think of them as generalizations of star charts and Chernoff faces.• A “mark” – a shape or form – that can be varied in n ways.

12 variables can be encoded – 4 angles, 4 lengths, 4 thicknesses

INFOVIS8803DV > SPRING 18

Examples of Multivariate Glyphs

INFOVIS8803DV > SPRING 18

Criticisms of Data Glyphs

One of the most common criticisms of data glyphs is that there is an implicit bias in most mappings, i.e., some attributes or relationships between attributes are easier to perceive than others.

• Perception-Based Bias• Proximity-Based Bias• Grouping-Based Bias

INFOVIS8803DV > SPRING 18

Let’s return to our beloved Spreadsheet …

What happens if we had 10s of columns and 100s of rows?How would you make the text more visual or symbolic?

INFOVIS8803DV > SPRING 18

Table Lens

Idea: • Make the text more visual and symbolic • Just leverage basic bar chart idea • Change quantitative values to bars • What about nominal data?

http://www.open-video.org/details.php?videoid=8304

https://www.youtube.com/watch?v=ZDY9YCYv7z8

INFOVIS8803DV > SPRING 18

LineUp

https://www.youtube.com/watch?v=iFqCBI4T8ks http://caleydo.org/tools/lineup/https://lineup.caleydoapp.org/

INFOVIS8803DV > SPRING 18

What about Categorical Data?

• How about multivariate categorical data?

• Students– Gender: Female, male– Eye color: Brown, blue, green, hazel– Hair color: Black, red, brown, blonde, gray– Home country: USA, China, Italy, India, ...

INFOVIS8803DV > SPRING 18

Mosaic Plot

INFOVIS8803DV > SPRING 18

Parallel Coordinates

Brushable http://bl.ocks.org/mbostock/1341021)Reorderable Axes http://bl.ocks.org/jasondavies/1341281http://www.parallelcoordinates.de/

INFOVIS8803DV > SPRING 18

Making Sense of Parallel Coordinates

INFOVIS8803DV > SPRING 18

Pixel Displays

• Represent data case or a variable as a pixel (or as a small glyph such as a circle) – Million or more per display– Seems to rely on use of color– Can pack lots in

• Challenge: What’s the layout? What does position mean?

INFOVIS8803DV > SPRING 18

One representation

• Grouping arrangement • One pixel per variable • Each data case has its own small rectangular icon • Plot out variables for data point in that icon using a grid or spiral

layout

INFOVIS8803DV > SPRING 18

Illustration

Levkowitz Vis ‘91

INFOVIS8803DV > SPRING 18

Unit Visualization

Sand Dance

https://sanddance.azurewebsites.net/BeachPartyApp/BeachPartyApp.htmlhttps://www.youtube.com/watch?v=15Hns2igiag

INFOVIS8803DV > SPRING 18

Unit Visualization

docplot

INFOVIS8803DV > SPRING 18

Another Metaphor

Dust-and-Magnet

• Iron bits (data) are drawn toward magnets (attributes) proportional to that data element’s value in that attribute

• Higher values attracted more strongly

• All magnets present on display affect position of all dust

• Individual power of magnets can be changed

• Dust’s color and size can connected to attributes as well

INFOVIS8803DV > SPRING 18

Another Metaphor

Dust-and-Magnet (Go Big)

https://www.cc.gatech.edu/gvu/ii/dnm/dustnmagnet-16Hub.mp4

INFOVIS8803DV > SPRING 18

Design Exercise

• Pair Up.• Visualize the data using any of the techniques we discussed in class.• Use colored pencils if needed.• Be prepared to discuss/present why you chose a particular technique

and how you encoded it.• Be creative.

Name XONE PS4 PC Gamespot Meta Critic New Price Used Price Genre

Titanfall 2 Yes Yes Yes 9 85 24.99 30.99 Action

Forza Horizon Yes No Yes 9 86 38.45 47.49 Sports

FIFA 17 Yes Yes Yes 9 84 34.89 33.95 Sports

Raven’s Cry Yes Yes Yes 1 27 11.99 0.05 Strategy

Gravity Rush No Yes No 8 -- 58.99 49.99 Action

Civilization VI No No Yes 9 88 41.40 19.99 Strategy

http://www.gamespot.com/

INFOVIS8803DV > SPRING 18

Set Data & Operations

• Large set of items, each can be in one or more sets• How do we visually represent the set membership?

INFOVIS8803DV > SPRING 18

Standard Technique

Venn Diagram

Contains all possible zones of overlap

INFOVIS8803DV > SPRING 18

http://www.conceptdraw.com/solution-park/diagram-venn

INFOVIS8803DV > SPRING 18

Euler Diagram

Only contains zones that exist

INFOVIS8803DV > SPRING 18

OnSet

https://www.cc.gatech.edu/gvu/ii/setvis/onset/index.html

INFOVIS8803DV > SPRING 18

Reminder -- PM2: Description

• Prepare a project description. • Create a two-page pdf document listing project members and

topic/data to be addressed. • Provide (initial ideas of) data source and format of the data

(spreadsheet, SQL database, etc.). • Provide a characterization of your target user. • Describe 3 or 4 questions/tasks that users of your system should be

able to investigate/answer. • Provide a sketch of your possible

• Submit a single pdf document on Canvas.

top related