Description

From a dashboard in your car to cutting-edge scientific papers, we extensively use visual representation of data. Thanks to the increasing amount of valuable data in every corner of our society, the visualization industry is growing rapidly and the visual analytics is becoming a crucial skill for knowledge workers. Effective analysis of data through visualization will become more and more crucial because it is almost impossible to understand big, messy data without any visual aid.

Why is visualization so powerful? Why some visualizations are more effective than others? What are the existing visualization methods? What are the current challenges in visualization?

In this course we will explore these questions by looking back the history of visualization, by analyzing and criticizing existing visualizations, and by creating our own visualization with data from our everyday life or your research problems. This course is both for the students who want to apply visualization techniques to their own work and for the students who want to develop better visualization algorithms and techniques.

Objectives

Students will learn why visualization matters and how to construct effective, meaningful, correct visualization.
Students will understand a wide array of visualization methods and be able to apply the techniques to their own work and data.

Basic information ↑

Time & Location: Informatics East 130
Tuesday & Thursday 9:30am–10:45am
First meeting: Jan. 10th, 2012
Instructor: Yong-Yeol Ahn (YY)
yyahn@indiana.edu
Office: Informatics East Room 316
Phone: (812) 856 2920
Office hours: Tuesday & Thursday 1pm–2:30pm, or by appointment
(you can email me or use MeetMe: http://doodle.com/yyahn)
AI: Abhik Seal
abseal@indiana.edu
Office: Lindley Hall Room 305
Announcements: All announcements will be sent through a mailing list (a Google group).
Textbook: Edward R. Tufte, The Visual Display of Quantitative Information (2nd Eds.)
Prerequisites: This course is open to graduate students as well as advanced undergraduate students. There are no formal prerequisites. It is, however, highly recommended to have basic working knowledge of programming (and strong willingness to learn new environments and languages) since we will need to manipulate data and figures. I will introduce visualization tools that do not require programming. Basic knowledge of statistics will be also helpful. Contact the instructor if you are uncertain about your background.

Syllabus ↑

This website contains all the information in the syllabus. You can simply print this website.

Schedule ↑

Date	Topics	Readings, Assignments, and Remarks
Why visualization matters?
1/10	Introduction and administrivia	Course Material: Wikipedia: 1854 Broad Street cholera outbreak GIS Analyses of Dr. Snow's Map Further readings: Economist: Worth a thousand words Steven Johnson, Ghost map Wikipedia: Voronoi diagram Matlab: voronoi
1/12	Why visualization?, Statistics 101, Histogram, The dark side of visualization	Course material: The Atlantic: Ending the infographic plague Flowing data: Fox News still makes awesome charts Bloomberg: Correlation or Causation? Further readings: Stephen Jay Gould, Full house Dawn Griffiths, Head First Statistics Assignment #1: Explore Gapminder and identify a relationship that is most fascinating to you. Submit a report explaining the data and the relationship with multiple figures Due date: 1/19, one week
1/17	Visualization examples, The darkside of visualization (cont'd)	Course material: US debts visualized New York Times: Budget puzzle: you fix the budget New York Times: How the deficit got this big Maplight Food desert locator gephi Google correlate Tableau Further readings: Darrell Huff and Irving Geis, How to lie with statistics Infovis wiki: Lie factor James Keirstead: Lie factors and airline seating Assignment #2: (i) browse news articles and find an example of bad visualizations we learned. Explain why it is bad and suggest how can we improve it (a sketch will help). (ii) Pick three visualization (bar chart, time series, etc) on mass media and calculate the lie factor (does not need to be super accurate). Due date: 1/26, one week + 2 days
Data
1/19	Data, Visualization examples, Data examples	Course material: Hans Rosling's 2006 TED talk Census data.gov ProPublica: Dollars for Docs Growth Rings Race and ethnicity Many Eyes Further readings: YYiki: Data scraping YYiki: Public datasets Propublica: Scraping for journalism: a guide for collecting data
Elements of Visualization
1/24	Data, Logarithmic scale	Course material: Hans Rosling's 2006 TED talk Power of ten The top ten worst graphs Further reading: Wikipedia: Logarithmic scale S Dehaene et al., Log or linear? Distinct intuitions of the number scale in Western and Amazonian indigene cultures, Science, 2008. Science news: Slide Rule Sense: Amazonian Indigenous Culture Demonstrates Universal Mapping Of Number Onto Space
1/26	Data, Logarithmic scale, 1D data	Course material: Google ngrams viewer Google ngrams datasets Khan academy: Logarithmic scale Khan academy: Vi and Sal Explore How We Think About Scale Khan academy: Vi and Sal Talk About the Mysteries of Benford's Law Khan academy: Benford's Law Explanation (Sequel to Mysteries of Benford's L Further reading: xkcd: http://xkcd.com/ngram-charts/ Khan academy: Richter scale Benford's Law and the Decreasing Reliability of Accounting Data for US Firms
1/31	1D data (contd.) Histogram Kernel Density Estimation	Course material: Wikipedia visualizations Histogram Probability distribution Probability density function Cumulative distiribution function Kernel density estimation Assignment #3: flip through scientific journals (e.g. Nature, Science, etc) to find one 2D plot (can be scatter plots, line charts, bar charts, etc) in linear scale and create rough sketches of them in log-scale (you can change only one axis to logarithmic scale or both). Then find one plot in log scale and create rough sketches in linear scale. For each case, explain why one scale (log or linear) is better than the other. Due date: 2/7 (one week)
2/2	KDE Histogram Cumulative plots	Course material: Histogram Cumulative distiribution function Kernel density estimation Kernel functions Further reading: Check out the posts on the mailing list Histogram: number of bins and width Stackexchange: stats: calculating optimal number of bins KDE: bandwidth selection optimal bandwidth code Bandwidth selection in kernel density estimation: oracle inequalities and adaptive minimax optimality Gnuplot info matplotlib ScipySuperpack Python resources rstudio
2/7	Box plots Processing D3.js Review	Course material: Hadley Wickham and Lisa Stryjewski, 40 years of boxplots Box Plot Processing Open processing D3.js Further reading: YYiki: processing YYiki: D3
Perception
2/9	Guest lecture: Kelly Caine	Psychophysics and Basic Human Visual Sensation and Perception
2/14	Log scale Eyes Visual acuity pixels and aliasing	Course material: C. Ware, Information Visualization: perception for design Aliasing Anti-aliasing Nyquist rate Further reading: Nyquist-Shannon sampling theorem
2/16	Visual angle Lightness Weber's law Steven's power law	Course material: Weber-Fechner law Stevens' power law YYiki: Optical illusion Checkershadow illusion Shaded diamonds Further reading: SS Stevens, The psychophysics of sensory function
2/21	Luminance, Brightness, Colors Preattentive processing	Course material: C. Ware, Information Visualization: perception for design xkcd, Color survey Further reading: Color blindness The day I saw Van Gogh's genius in a new light
Gues lectures
2/23	Guest lecture: Angela Zoss	Data Processing and Analysis Kandel, Heer, Plaisant, et al. (2011). Research directions in data wrangling: Visualizations and transformations for usable and credible data. Information Visualization, 10(4), 271–288. link Reshef, D. N., Reshef, Y. A., Finucane, H. K., Grossman, S. R., McVean, G., Turnbaugh, P. J., Lander, E. S., Mitzenmacher, M., & Sabeti, P. C. (2011). Detecting novel associations in large data sets. Science, 334, 1518. link Google Refine (http://code.google.com/p/google-refine/) Pig (http://research.yahoo.com/node/90) R (http://www.r-project.org/) Mr. Data Converter (http://shancarter.com/data_converter/) Mr. People (http://people.ericson.net/) Data Science Toolkit: various conversion and lookup tools (http://www.datasciencetoolkit.org/)
2/28	Guest lecture: David Crandall	Reconstructing the world from social photo-sharing websites http://vision.soic.indiana.edu/disco http://www.cs.cornell.edu/~crandall/photomap/
3/1	Guest lecture: Angela Zoss	Comparison techniques Football, Physics, and Symmetry Envisioning information by E.R. Tufte The words they used Flowing data:How to Make a Heatmap – a Quick and Easy Solution Last.fm scrobbles as calenar heat map Envision digital library project TIBCO Spotfire DecisionSite Enterprise Analytics Product Suite GRIDL Algorithm Dynamic pairs plot
3/6	Project proposal presentation
Maps
3/8	Maps	Course material: YYiki: Geovisualization The earliest known map Wikipedia: Perspective (graphical) Wikipedia: Map, Wikipedia: Cartography Ask ET: Flowlines The WebGL Globe xkcd: Map projections Wikipedia: Map projections The True Size of Africa Further reading: The golden age of visualization, by Juan C. Dürsteler Wikipedia: Edmond Halley Wikipedia: Charles Joseph Minard LASCAUX: THE MYSTERIOUS PATRON
3/13	No class (spring recess)
3/15	No class (spring recess)
3/20	Mercator projection Geocoding Cartogram	Course material: Wikipedia: Mercator projection Wikipedia: Choropleth map YYiki: Geocoding YYiki: Cartogram Further reading: VF Rickey and PM Tuchinsky, An Application of Geography to Mathematics: History of the Integral of the Secant, Mathematics Magazine, 1980 J. Alexander, Loxodromes: a rhumb way to go, Mathematics Magazine, 2004 Rhumb line
Historical visualization
3/22	Lily Library	Course material: Ptolemy, 2nd cent. [Geographia. Latin] Cosmographia. Ulm : Lienhart Holle, 16 July 1482. G1005 1482 vault Abraham Ortelius. Theatrum orbis terrarum. Originally published: Antwerp: Gillis Coppens van Diest, 1570. G1006 .T37 F8 1570 Dutch East India Company Coastal Profiles Boxer II mss vault Breydenbach, Bernhard von. Peregrinationes ad Terram Sanctam. Peter Schoffer, Mainz, 1486. DS 106.B7 vault Louis Bretez. Plan de Paris, Commencé l’Année 1734, et Gravé sous les ordres de Messire Michel Etienne Turgot… (Paris, 1739) G1844 .P3 B844 vault Euclid, Elements (the first english edition, 1570) William Playfair. For the use of the enemies of England: A real statement of the finances and resources of Great Britain; Illustrated by two copper-plate charts. London: Printed by C. Whittingham …, 1796. HG938 .P7 Mortality of the British army : at home, at home [sic] and abroad, and during the Russian war, as compared with the mortality of the civil population in England. Published: London : Printed by Harrison and Sons, 1858. DK215 .G7 Marey, Dr. E.J. Physiologie médicale de la Circulation du sang basée sur l’étude grahique des mouvements du coeur et du pouls artériel … Paris: Adrien Delahaye, 1863. RC683.5 .C26 M326 Serio-comic War Map for the Year 1877 / by F.W. Rose. London: G.W. Bacon, [1877]. Political cartoon map in five colors. 51 x 65 cm. London Low Life DA676 Oversize folder 4, no. 1 Marey, E. –J. Physiologie du mouvement. Le Vol des oiseaux. Paris: G. Masson, 1890. QL698 .M32 To secure these rights: The report on the President’s committee on Civil Rights. NY: Simon and Schuster, 1947. PS3531 .L7 Box 15 no. 10 Assignment #4: check out oncourse (Due: 4/3)
Graphs and Trees
3/27	Progress report 1
3/27	Review: Lily library Cartogram Tree Network Tree map	Course material: Wikipedia: Geography (Ptolemy) Guardian: Florence Nightingale, datajournalist: information has always been beautiful Wikipedia: E.-J. Marey YYiki: Cartogram FlowingData: How to Make a US County Thematic Map Using Free Tools Wikipedia: Treemapping Further reading: the Pioneers: Étienne-Jules Marey BBC: The beauty of maps: Serio-comic war map, 1877 M. O. Ward, G. Grinstein, D. Keim, Interactie Data Visualization
3/29	Treemap Sunburst plot Interpolation Bezier curve Planar graph	Course material: M. O. Ward, G. Grinstein, D. Keim, Interactie Data Visualization John Stasko: SunBurst D3.js: sunburst Wikipedia: Interpolation Wikpedia: Multivariate interpolation Geometry in action: Interpolation and extrapolation Wikipedia: Regression analysis Wikipedia: Bezier curve Wikipedia: Planar Graph Further reading: Visual complexity Pat Hanrahan, To draw a tree Tamara Munzner et al., TreeJuxtaposer: scalable tree comparison using Focus+Context with guaranteed visibility
4/3	Bézier curves and splines Force-directed layouts Hierarchy Networks and Matrices	Course material: bezier.java Mike Bostock, d3.js: Data-driven documents flare demo Protovis examples Fighting (with) Hierarchies – Presentation Techniques I Wordtree: Visualizations : Obama Nobel Remarks World Wikipedia: Dendrogram Bipartite networks: some examples Further reading: Paul Haeberli, dynadraw Sara L. Su et al., Simulating artistic brushstrokes using interval splines Thomas Strothotte and Stefan Schlechtweg Non-photorealistic computer graphics Bruch Gooch and Amy Gooch, Non-photorealistic rendering Aaron Hertzmann, Algorithms for Rendering in Artistic Styles Martin Wattenberg and Fernanda B. Viégas, The word tree, an interactive visual concordance Mohammad Ghoniem et al., A comparison of the readability of graphs using node-link and matrix-based representations
4/5	Hierarchical edge bundling Force-directed edge bundling Exploratoy network visualization convex hull	Course material: d3 wiki Danny Holten Hierarchical edge bundles: visualization of adjacency relations in hierarchical data Force-directed edge bundling for graph visualization Link community visualization Wikipedia: Convex hull Community Structure in Multi-Scale Transportation Networks Further readling: Beautiful visualization
4/10	Progress report 2
4/10	Network visualization tools Text visualization	Course material: Gephi Cytoscape Spato visual explorer Wordle Neoformix: State of the Union 2011 blprnt: Tokyo \| Cairo: comparing Obama's foreign policy speeches Jacob Harris, Word clouds considered harmful NYT, A deadly day in Baghdad NYT, Choice words BBC, State of the Union: The rise of 'we' NYT, The 2007 State of the Union Address Ben Fry, On the origin of species: the preservation of favoured traces Further reading: Stephen G. Eick et al., Seesoft: a tool for visualizing line oriented software statistics Fernanda Viégas and Martin Wattenberg, Tag clouds and the case for vernacular visualization
Texts
4/12	Text visualization	Course material: Spato visual explorer History flow Bible cross references visualization The shape of song Many eyes: Word tree: I have a dream Many eyes: Phrase nets We feel fine NYT, Naming names Further reading: Fernanda B. Viégas et al., Studying cooperation and conflict between authors with history flow visualizations Koji Yatani et al., Analysis of Adjective-noun Word Pair Extraction Methods for Online Review Summarization Frank vsn Ham et al., Mapping text with phrase nets YYiki: Sentiment analysis IU: B651: Natural language processing Stanford NLP class
Interaction
4/17	Graph visualization Interaction	Course material: NodeXL sigma.js Jeffrey Heer and Ben Shneiderman, Interactive dynamics for visual analysis New York Times: The Jobless Rate for People Like You
4/19	Course Review	Course material: Journalism in the age of data
4/24	Project presentation
4/26	Project presentation
4/30	Project paper due

Projects ↑

Objectives

By carrying out a visualization project, you will get a hands-on experience of:

finding and analyzing data that you care about,
applying various visualization methods and understanding why and when to apply certain visualization techniques,
identifying visualization problems, develop a new visualization methods, and implement softwares that people can use.

You can either choose to write an in-depth article about a topic of your choice with data analysis and various visualization techniques or develop a new visualization algorithm or software.

Deliverables

The deliverables are

Project proposal (Due: 3/2)

An one or two page document that contains

Project title
Team members
Project description
Relevant prior work

Proposal presentation (Due: 3/5, presentation on 3/6)

We will follow the Ignite format. You should have 20 slides and each slide will auto-advance every 15 seconds. You should submit the slides the day before the class. It should address

Motivation (why is it interesting?)
Relevant prior work
Potential contribution (why and how does your project differ from prior work?)
Game plan
Division of labor (who will do what?)
Current progress (if any)

Progress report (#1 Due: 3/27, #2 Due: 4/10)

Think of it as a draft for the final report, which should contains failures and successes in the course of your project. It should address: what you have tried, what didn't work, what did work, and what you have learned from it. There is no specific format. Try to document details that you have done. For instance, if you try to learn D3.js, summarize the things that you learned, show screenshots or codes of the visualizations that you played with, in learning it. Document your design sketches, brainstorming ideas, failed plots, etc.

Final report (4/28)

Prepare two documents: one is a polished document that resembles an academic paper with the following elements.

Abstract: a short summary of the project your achievements
Introduction: motivation, relevant prior work, description of your approaches
Methods: describe the methods used in the paper
Results: Describe your results in detail with visualization techniques.
Discussion and Conclusion: describe impacts, caveats, limitations, potential extensions, and what you have learned.
Author contributions: describe how each one of you contributed to the project.

In the other document, document your *sweat* on the project. Record your failures and successes and desribe what you have learned about visualization through your experiences.

Final presentation (4/24, 4/26)

We will follow the Ignite format but this time 10 minues for each team. You should have 40 slides and each slide will auto-advance every 15 seconds. You should submit the slides by the night of 4/23.

Resources

Policies ↑

Class policies

All announcements will be sent via email. Students are responsible for reading each announcement in detail.
Students need to read all the assigned readings prior to the class.
Students have the responsibility of backing up all their data and code. Today is International Backup Awareness Day! Use a backup drive, dropbox, or whatever service you find it useful. For the students who are comfortable with commandlines, I highly recommend using a version control system, especially with hosting services such as github (IU provides a firewalled github), bitbucket, etc.
Please contact the instructor if you have a disability that require some arrangements so that appropriate arrangements can be made.

Academic integrity

The principles of academic honesty and ethics will be enforced. Any cases of academic misconduct (cheating, fabrication, plagiarism, etc) will be thoroughly investigated and immediately reported to the School and the Dean of Students. You should actively discuss with others, but you should write your own report. Credit all the sources (discussion with other students, used softwares, etc).

IU Code of Student Rights, Responsibilities, and Conduct

Grading policy

Project: 40%
Assignments: 40%
Attendance & Class participation 20%

Resources ↑

Softwares, libraries, and data

General

Tableau
Many eyes
D3.js
Processing (Processing examples and codes) and Processing.js
Cinder
The Visualization Toolkit
R and ggplot2 (more resources can be found in my wiki)
Simile
Javascript infovis toolkit
gnuplot
Paper.js

I400 / H400 / I590: Visual Analytics

Spring 2012

Description

Objectives

Basic information ↑

Syllabus ↑

Schedule ↑

Projects ↑

Objectives

Deliverables

Project proposal (Due: 3/2)

Proposal presentation (Due: 3/5, presentation on 3/6)

Progress report (#1 Due: 3/27, #2 Due: 4/10)

Final report (4/28)

Final presentation (4/24, 4/26)

Resources

Policies ↑

Class policies

Academic integrity

Grading policy

Resources ↑

Links

Softwares, libraries, and data

General

Network visualization

Geo-visualization

Books

I400 / H400 / I590: Visual Analytics

Spring 2012

Description

Objectives

Basic information ↑

Syllabus ↑

Schedule ↑

Projects↑

Objectives

Deliverables

Project proposal (Due: 3/2)

Proposal presentation (Due: 3/5, presentation on 3/6)

Progress report (#1 Due: 3/27, #2 Due: 4/10)

Final report (4/28)

Final presentation (4/24, 4/26)

Resources

Policies ↑

Class policies

Academic integrity

Grading policy

Resources ↑

Links

Softwares, libraries, and data

General

Network visualization

Geo-visualization

Books

Projects ↑