Think New Shapes Rado (2004)

What is Visualization Really for? Min Chen Professor of Scientific Visualization

Oxford e-Research Centre University of Oxford

[email protected] RIVIC Graduate School, 10-11 April 2013

Word Occurrences in the Paper Titles of VisWeek 2010 (InfoVis and SciVis)

Outline

Tag Cloud created using IBM Many Eyes

1. What is visualization and

what is it really for? 2. How does visualization do that? 3. What are the challenging problems in visualization?

1. What is visualization?

Image from: http://www.positive-thinking-for-you.com/

Types of Visualization (by Input Data)          

Textual Data Network Data Tubular Data Software Volume Data Vector Field Tensor Field Geo-information Bio-information ...

Visually More Realistic

Faster & More Interactive

More Illustrative & Expressive

Better Accuracy

More Information about Data

Information about a in b

local statistical complexity

relationship between past and future

Larger Data Sets

63 Terabyte

Larger User Base

Larger Problem Scale

Claude E. Shannon (1916-2001)

Theory of Visualization

p(x, y, z) = p(x) p(y|x) p(z|y) p(x)

p(y|x)

X

Process 1

p(z|y) Y

I (X; Y)

Process 2 I (Y; Z)

X

Process 1

interaction U2 Y

Process 2

information loss: 25.8%

256

Visualization Can Break the Conditions of Data Processing Inequality interaction U1

Chen and Jänicke, TVCG, 2010

Z

Z

192

192

128

128

64

64

0

0 0

8 16 24 32 40 48 56 64

0

(a) evenly distributed p

domain knowledge about X

information loss: 22.6%

256 192

X

Process 1

Y

Process 2

Z

information loss: 25.0%

256

(b) unevenly distributed p

28

128

26

64

24

I ( X ;Y ) ≥ I ( X ; Z )

8 16 24 32 40 48 56 64

information loss: 0%

22

20

0 0

8 16 24 32 40 48 56 64

(c) 4 regional mappings

0

8 16 24 32 40 48 56 64

(d) logarithmic plot

What is visualization really for? 

What made me ask this question?

OeRC mini powerwall

ISIC powerwall

Conventional Definitions 

“The goal of visualization in computing is to gain insight by using our visual machinery” [McCormick et al. 1987]



“... a method for seeing the unseen ... fosters profound and unexpected insights” [McCormick et al. 1987]



“... maximize human understanding and communication” [Owen 1999]



“... gain understanding and insight into the data. ... promote a deeper level of understanding ... foster new insight ...” [Earnshaw and Wiseman 1992]



“... to gain insight into an information space ...” [Senay and Ignatius 1990]



“... to amplify cognition” [Card et al. 1999]



“... graphics can be more precise and revealing than conventional statistical computations” [Tufte 2001]



“Information visualization helps think.” [Few 2009]



“... to assist humans in solving problems” [Purchase et al. 2008]



“... unveiling of the underlying structure” [Berkeley 2010]

The Chart-Junk Debate 

Nigel Holmes, 1984



Edward Tufte, 2001



Scott Bateman et al. 2010



Stephen Few, 2011a 

“At best we can treat the findings as suggestive of what might be true, but not conclusive.”



Jessica Hullman et al. 2011



Stephen Few, 2011b 



“If they’re wrong, however, which indeed they are, their claim could do great harm.”

Rita Borgo, et al. 2012

Enabling Tool 

Save time



Spot patterns  

£11.00 £10.50 £10.00



Share Price of Company X



£9.50



£9.00 £8.50 £8.00 £7.50 £7.00 £6.50 £6.00 £5.50 £5.00



External memorisation



Stimulate hypotheses



Visually “evaluate” hypotheses 

£4.50

2002

2003

2004

distribution clusters anomalies correlation ...

save time (repeat)

2002

MIN

AVG

MAX

January February March April May June July August September October November December

5.45 5.55 5.56 5.54 5.58 6.05 6.75 x.xx x.xx x.xx x.xx x.xx

5.56 5.82 5.70 6.01 6.03 6.23 6.99 x.xx x.xx x.xx x.xx x.xx

5.65 5.93 5.91 6.75 6.81 7.12 7.31 x.xx x.xx x.xx x.xx x.xx

2003

MIN

AVG

MAX

January February March April May June July August September October November December

x.xx x.xx x.xx x.xx x.xx x.xx x.xx x.xx x.xx x.xx x.xx x.xx

x.xx x.xx x.xx x.xx x.xx x.xx x.xx x.xx x.xx x.xx x.xx x.xx

x.xx x.xx x.xx x.xx x.xx x.xx x.xx x.xx x.xx x.xx x.xx x.xx

2003

MIN

AVG

MAX

January February ...

x.xx x.xx ...

x.xx x.xx ...

x.xx x.xx ...

Enabling Tool 

Save time



Spot patterns 

distribution



clusters



anomalies



correlation



...



External memorization



Stimulate hypotheses



Visually “evaluate” hypotheses 

save time (repeat)

Enabling Tool £11.00 £10.50

Share Price of Company X

£10.00 £9.50 £9.00

September December February November December January October August April March June July May 2004 2002 2003 2002 low: £7.50 £7.00 £9.30 £5.40 £5.60 £5.50 £6.00 £6.50 £6.70 £6.90 £7.80 £7.70 £7.40 £7.90 £8.50 £8.80 £9.40 £9.20 £9.10 £8.40 £7.20 £7.30 £8.70 £8.20 £8.00 high: £8.30 £7.40 £9.60 £5.70 £5.90 £7.10 £7.20 £8.00 £8.20 £8.40 £8.50 £8.90 £9.40 £9.70 £9.50 £7.70 £7.90 £8.10 £9.30 £9.00 £8.60 end: £7.70 £7.20 £9.40 £5.60 £5.80 £5.70 £6.00 £6.20 £6.60 £7.00 £7.40 £7.10 £7.90 £8.20 £8.50 £8.80 £9.50 £9.30 £9.20 £9.00 £7.60 £7.80 £8.00 £8.60 £8.30



Save time



Spot patterns 

distribution



clusters



anomalies

£7.00



correlation

£6.50



...

£8.50 £8.00 £7.50

£6.00



External memorization

£5.00



Stimulate hypotheses

£4.50



Visually “evaluate” hypotheses

£5.50



save time (repeat)

What is visualization really for?

Save Time

2. How Does Visualization Save Time?

How does visualization save time?



Making observation: Overview 

e.g., bringing time and attributes together

Visualizing Glacier Movement

10 years 200+ glaciers

Drocourt et al., CGF, 2011

Complexity Plots 

Visualization for CS

Thiyagalingam et al., CGF, 2013

How does visualization save time?



Making observation: Overview 



e.g., bringing time and attributes together

Making observation: Omission 

e.g., seeing “time” without using “time”

Three Similar Cases in CAVIAR Datasets

LeftBag

LeftBag_PickUp

LeftBox

Three Visualizations Chen et al., TVCG, 2006

LeftBag

LeftBag_PickUp

LeftBox

VideoPerpetuoGram (VPG) 

Can we make video visualization as useable as Electrocardiogram (ECG) and Seismographs?

Botchen et al., TVCG, 2008

Snooker Training Höferlin et al., CGF, 2010 Parry et al., TVCG, 2011

How does visualization save time?



Making observation: Overview 



Making observation: Omission 



e.g., bringing time and attributes together

e.g., seeing “time” without using “time”

External memorization 

e.g., making the brain do more useful work

Example: Match Visualization 

Real-time or offline annotation results in a huge spreadsheet of events

Legg et al., CGF, 2012

Poetry Visualization (with Utah) 

Supporting close reading

Abdul-Rahman, et al., CGF, 2013

How to display many measurements at the same time? Duffy et al., under review, 2013

How much information can glyphs encode? SLD: Straight Line Direction (shape orientation)

MAD: Mean Angular Displacement (angle)

HP: Head Position (center position) HW: Head Width (shape width) HL: Head Length (shape length) HR: Head Rotation (shape orientation)

BCF: Beat-Cross Frequency (length/circumference) Uncertainty (colour)

LIN*: VSL/VCL

ALH: Amplitude of Lateral Head Displacement (length)

WOB*: VAP/VCL STR*: VSL/VAP

V0: Zero Velocity (fixed radius)

VSL: Straight Line Velocity (radius and colour) VAP: Average Path Velocity (radius) VCL: Curvilinear Velocity (radius)

FCA: Change-in-Angle of Filament

FTT: Total Torque of Filament (thickness) FAS: Asymmetry of Filament (angular displacement)

FTA: Total Projected Arclength of Filament

Can we see time (video) without using time (animation)?

How does visualization save time?



Making observation: Overview 



Making observation: Omission 



e.g., seeing “time” without using “time”

External memorization 



e.g., bringing time and attributes together

e.g., making the brain do more useful work

Hypothesis generation and evaluation 

e.g., using intuition, experience and knowledge

Example: Facial Dynamics 

Expression Recognition   



Data   



Humans are very good at Machine vision is far behind Limited understanding Video Feature changes Time series

Challenges   

A lot of features A lot of ways of measuring features Non-uniform temporal behavior

Tam et al., CGF, 2011

Parallel Coordinates 

Multi-dimensional data visualization Y

Y

X

X

Interactive Visualization: Outliers

Interactive Visualization: Formulating Decisions

How does visualization save time?



Making observation: Overview 



Making observation: Omission 



e.g., making the brain do more useful work

Hypothesis generation and evaluation 



e.g., seeing “time” without using “time”

External memorization 



e.g., bringing time and attributes together

e.g., using intuition, experience and knowledge

Making pixels do the work 

e.g., visual multiplexing

Visual Multiplexing





10 different ways of delivering multiple pieces of information associated with specific location. Application case study in Cardiovascular Magnetic Resonance Imaging

Walton et al., under review, 2013

How does visualization save time?



Making observation: Overview 



Making observation: Omission 



e.g., using intuition, experience and knowledge

Making pixels do the work 



e.g., making the brain do more useful work

Hypothesis generation and evaluation 



e.g., seeing “time” without using “time”

External memorization 



e.g., bringing time and attributes together

e.g., visual multiplexing

Effective Communication 

e.g., getting the message across

Visual Embellishment







Positive impact on memory Negative impact on visual search Likely positive impact on concept grasping

Borgo et al., TVCG, 2012

3. What are the challenging problems?

Data Deluge ... peta (240), exa (250), ... Time-dependent Often unstructured Usually with uncertainty Interrelated

Alexander Graham Bell (1847-1922)

Telephone



In the 1870s, Bell travelled around to give demos ‘in concert halls, where full orchestras and choruses played “America” and “Auld Lnag Syne into his gadgetry.’



Around 1880, Queen Victoria installed a pair of telephones at Winsor and Buckingham Palace

Primary source: J. Gleick, book, 2012

Comments on Telephone in 1870s 

In 1878, in the USA, Theodore Vail quit the Post Office Department and joined the Bell Telephone Company. A colleague commented: 

“I can scarcely believe that a man of your sound judgement ... should throw it up for a d...d old Yankee notion (a piece of wire with two Texan steer horns attached to the end, with an arrangement to make the concern blate like a calf) called a telephone.”

Primary source: J. Gleick, book, 2012

Comments on Telephone in 1870s 

In 1879, in England, the chief engineer of the General Post Office, William Preece, reported to Parliament: 

“I fancy the descriptions we get of its use in America are a little exaggerated, ... Here we have a superabundance of messengers, errand boys and things of that kind. ... I have one in my office, but more for show. If I want to send a message  I use a sounder or employ a boy to take it.”

Primary source: J. Gleick, book, 2012

“Mr. Information, come here. I want to see you.”

Visualization is for saving time by using more efficient and effective means in the process of discovery and communication of information and knowledge

Visualization 

>1000 years ago 



First known line graph

1700s 

Statistics Graphics  William Playfair 

1963 

Sketchpad  Ivan Sutherland 

1989 



1995 



First symposium on Information Visualization

2006 



First workshop on Scientific Visualization

First symposium on Visual Analytics

2013 

VisWeek is renamed as VIS

Challenge One: Big Data  

What do we want to know? Causality discovery

Chen et al., IEEE Computer, 2011

openconnectomeproject.org 1000 images, each with 112,500 x 87,500 pixels

Challenge Two: Theory of Visualization Jänicke et al., CGF, 2011

    

Measurements Explanation Quantitative Laws Models Prediction Chen and Jänicke, TVCG, 2010





Information Theories in Communication Measurements and Hypothesized Models in Cognitive Sciences

Challenge Three: Visualizing Time without Using “Time”

   

Time series Video Events Daniel and Chen, IEEE Vis, 2003 ... Botchen et al., TVCG, 2008 Hoeferlin et al., CGF, 2010 Parry et al., TVCG, 2010 Jänicke et al., CGF, 2010 Duffy et al., under review, 2013

Challenge Four: Automation, “with Humans in the Loop”  

Can visualization be automatically generated? Can visual designs be guided by computation? Maguire et al., TVCG, 2012 Gilson et al., CGF, 2008

Think New Shapes Rado (2004)

How to Save Time? Think new shapes!

Acknowledgement University of Oxford  Alfie Abdul-Rahman  Kai Berger  Brian Duffy  Saiful Khan  Eamonn Maguire  Karl Proctor  Jeyan Thiyagalingam  Simon Walton 

Colleagues in OeRC, OCCAM, ...

Swansea Rita Borgo  Phil W. Grant  Iwan Griffiths  Mark W. Jones  Bob Laramee  Adrian Morris  Tavi Murray  Irene Reppa  Kilian Scharrer  Ian Thornton 



ROs and PhDs (below)

Past PhDs and ROs:

 

               

C.-Y. Wang (PhD, 1989-1992) Mark W. Jones (PhD, 1991-1994) Abdula Haji Tablib (PhD, 1990-1994) Mike Bews (PhD, 1992-1996) Malcolm Price (MPhil, 1997-1998) Adrain Leu (PhD, 1996-1999) Simon Michael (PhD, 1996-1999) Steve Treavett (PhD, 1997-2000) Mark Kiddell (RA, 1999-2001) Ben Smith (TCA, 1999-2001) S.-S. Hong (PhD, 1998-2002) Abdul Haji-Ismail (PhD, 1998-2002) H.-L. Zhou (MPhil, 2000-2002) Andrew S. Winter (PhD, 1999-2002) David Rogeman (PhD, 1999-2003) Paul Adams (TCA, 2002-2004)

               

Stuttgart  Tom Ertl  Daniel Weiskopf  Ralf Botchen ... Rutgers  Deborah Silver  Carlos Correa Purdue (VACCINE)  David Ebert Heidelberger  Heike Jänicke

Tim Lewis (RA, 2004-2005) Gareth Daniel (PhD, 2001-2004) David P. Clark (PhD, 2001-2005) Dave Bown (RA, 2005) Ann Smith (PhD, RA, 2001-2006) Siti Z. Zainal Abdin (PhD, 2003-2007) Alfie Abdul Rahman (PhD, RA, 2004-7) Joanna Gooch (PhD, 2004-2007) Shoukat Islam (PhD, RA, 2004-2009) David Chisnall (PhD, RA, 2005-2008) Phil Roberts (RA, 2005-2008) Rudy R. Hashim (PhD, 2005-2008) Dan Hubball (MPhil, 2007-2008) Owen Gilson (PhD, 2006-2009) Lindsey Clarke (PhD, 2007-2010) Heike Jänicke (RO, 2009-2010) Farhan Mohamed (PhD, 2008-) Ed Grundy (PhD, 2009-)

        

Utah Chris Johnson, Kate Coles, Julie Lein, Miriah Meyer  Chuck Hansen Cardiff  Andrew Aubrey  Dave Marshall  Paul Rosin  Gary Tam RIVIC  Nigel John  Ralph Martin  Reyer Zwiggelaar 

Rita Borgo (2009-2011) Hui Fang (2009-2011) Yoann Drocourt (PhD, 2010-2011) Karl Proctor (PhD, 2009-2011) Andrew Ryan (PhD, 2010-2011) Phil Legg (RO, 2010-2011) David Chung (PhD, RA, 2010-2011) Matthew Parry (MPhil, RA, 2010-2011) Richard M. Jiang (RO, 2010-2011)

What is Visualization Really for?

information loss: 22.6% information loss: 0%. Visualization Can Break the. Conditions of Data. Processing Inequality. Claude E. Shannon (1916-2001). Chen and Jänicke, TVCG, 2010 ... to gain insight into an information space ...” [Senay and Ignatius 1990] ..... Graham Bell (1847-1922). Primary source: J. Gleick, book, 2012 ...

5MB Sizes 3 Downloads 208 Views

Recommend Documents

what is marriage, really? - Lutheran Press
the publication history of this work, along with the social and theological ...... as sinful, you should brand the other one likewise. ... children, as one picks an apple.

what is marriage, really? - Lutheran Press
Swan logo is a LifeART image © 2013 Lippincott Williams &. Wilkins. All rights ... The first is to recognize that marriage is ...... children, as one picks an apple.

What neuroeconomics does really mean?
namely the ventromedial prefrontal (VM) and the amygdala have much more difficulties to .... is delayed for a longer term (Mac Lure and alii, 2004). So reasoning ...