CIS 4930/6930-002
DATA VISUALIZATION
Introduction to Visualization
Paul Rosen
Assistant Professor
University of South Florida
(slides adapted from Miriah Meyer)
• WHY STUDY VISUALIZATION?
d ta
data
INDUSTRIAL REVOLUTION OF DATA
Joe Hellerstein,UC Berkley,2008
•HOW MUCH DATA IS THERE?
2010: 1.2 zettabytes
•
2013: 4.4 zettabytes
•
2020: ~40 zettabytes
•
2.5 quintillion bytes / day (2.5 EB)
•
zettabyte ~=
•
1,000,000,000,000,000,000,000 or 1021
(200x all words ever spoken by humans)
•
•The ability to take data—to be able to understand it,
to process it, to extract value from it, to visualize it, to
communicate it—that’s going to be a hugely
important skill in the next decades...
Because now we really do have essentially free and
•
ubiquitous data. So the complimentary scarce factor is
the ability to understand that data and extract value
from it.
Google’s Chief Economist, Hal Varian, 2009
•
• http://flowingdata.com/2009/02/25/googles-chief-economist-hal-varian-on-statistics-and-data/
• Big Data: Experts say new forms of information analysis will help
people be more nimble and adaptive, but worry over humans'
capacity to understand and use these new tools well
Tech experts believe the vast quantities of data that humans and machines
•
will be creating by the year 2020 could enhance productivity, improve
organizational transparency, and expand the frontier of the "knowable future."
But they worry about "humanity's dashboard" beingin government and
corporate hands and they are anxious about people's ability to analyze it wisely.
Janna Quitney Anderson, Elon University
•
• lee Rainie, Pew Research Center's Internet & American Life Project
July 20,2012 •
• http://www.pewinternet.org/wp-content/uploads/sites/9/media/Files/Reports/2012/PIP_Future_of_Internet_2012_Big_Data.pdf
•DATA SCIENCE TOP
JOB IN 2017 & 2018
ACCORDING TO
GLASSDOOR.COM
• WHY DO WE CREATE VISUALIZATIONS?
answer questions
•
generate hypotheses
•
make decisions
•
see data in context
•
expand memory
•
• support computational analysis
find patterns •
tell a story •
inspire •
• VISUALIZATION GOALS
record information
•
• analyze data to support reasoning
confirm hypotheses
•
communicate ideas to others
•
Record Information
• http://citrinitas.com/history_of_viscom
• http://citrinitas.com/history_of_viscom
• Visual Thinking Collection, Dave Grey
• Leonardo daVinci 1485
• Galileo 1610
• E.J.Muybridge 1878
ANALYSIS TO SUPPORT REASONING
•ANALYSIS
TO REVEAL
PATTERNS
Source: New York Times
http://www.nytimes.com/2014/08/16/upshot/mapping-migration-in-the-united-states-since-1900.html?_r=0
THE CHALLENGER
Confirm
DISASTER, 1986 Hypotheses
• source:Wikipedia
source: E. Tufte
source: E. Tufte
CONFIRM HYPOTHESES
London Cholera
Outbreak
John Snow 1854
COMMUNICATE IDEAS
Florence Nightingale 1856
Joseph Minard 1861
• WHY DOES VISUALIZATION WORK?
13
•WHY DOES
VISUALIZATION WORK?
• cognition is limited
• WHY DOES VISUALIZATION WORK?
cognition is limited
•
memory is limited
•
34
•
calculation exercise . . . x28
•
• 952
79
•
calculation exercise . . . x16
•
• 1264
•34
calculation exercise . . . x28
•
• HOW VISUALIZATION WORKS
VISUALIZATION
•
• uses perception to point
out interesting things
MTHIVLWYADCEQGHKILKMTWYN
•
ARDCAIREQGHLVKMFPSTWYARN
GFPSVCEILQGKMFPSNDRCEQDI
PSGHLMFHKMVPSTWYACEQTWRN
MTHIVLWYADCEQGHKILKMTWYN
•
ARDCAIREQGHLVKMFPSTWYARN
GFPSVCEILQGKMFPSNDRCEQDI
PSGHLMFHKMVPSTWYACEQTWRN
• VISUALIZATION
• uses pictures to enhance
working memory
15 19 60
33 11 75
57 34 79
18 51 92
73 22 13
71 60 22 • GIVEN THESE NUMBERS . . .
17 10 68 • . . . what number appears most often?
73 18 55
65 46 29
60 73 22
46 92 97
10 58 46
57 17 83
26 99 33
88 92 60
91 29 57
96 12 47
100
80
• GIVEN THESE NUMBERS . . .
60 • . . . what number appears most often?
40
20
0
• QUERY EXERCISE . . .
TRIGLYCERIDE LEVEL
•QUESTION: Which gender and income
level shows a different effect of age on
triglyceride levels?
TRIGLYCERIDE LEVEL
•QUESTION: Which gender and income
level shows a different effect of age on
triglyceride levels?
The goal of this course is to introduce
•
students to the principles, methods, and
techniques for effective visual analysis of data
• We will discuss visualization techniques for a
broad range of data types.
• You will gain experience using visual analysis
tools, as well as in developing your own
interactive visualization tools.
• FOUNDATIONS
design •
perception
•
data and tasks
•
visual encoding
•
interaction
•
multiple views
•
• filtering and aggregation
• TECHNIQUES
•tabular data
• graphs and trees
text
•
sets
•
maps
•
• ADVANCED TOPICS
High dimensional data
•
Vis + Stats
•
• Vis + Machine Learning
Vis + Topology
•
• GROUND RULES
• be respectful of everyone’s time
come to class prepared
•
critique ideas, not people
•
discuss ideas together, do the
•
assignments on your own
be engaged!
•
Recommended
Watching
• RECOMMENDED READING
• Visualization Analysis & Design: Chapter 1 (pp. 1-19)
