About This Talk � What is information visualization � Principles of graphical excellence Introduction to � Principles of integrity Information Visualization � Some visualization techniques � References Kai Li � E.R. Tufte, The Visual Display of Quantitative Information , Computer Science Department Graphics Press, 1983. Princeton University � S.K. Card, J.D. Mackinlay, and B. Shneiderman, Information Visualization: Using Vision to Think, Morgan Kaufmann Publishers, 1999. 2 What is Information Visualization? Power of Visualization � Visualization : From Princeton CS Department to “The action or fact of visualizing; the power or process of forming Rutgers’ CS Department: a mental picture or vision of something not actually present to the sight; a picture thus formed.” (Oxford English Dictionary) � Start out going South on OLDEN ST toward PROSPECT AVE. � � Information visualization : Turn RIGHT onto PROSPECT AVE. � “Transformation of the symbolic into the geometric” (McCormick Turn LEFT onto WASHINGTON RD/ CR-526/ CR-571. et al., 1987) � Turn RIGHT. � Information visualization : � Turn LEFT onto US-1 N/ “... finding the artificial memory that best supports our natural BRUNSWICK PIKE. Continue to means of perception.'‘ (Bertin, 1983) follow US-1 N. � Information visualization: � Merge onto NJ-18 N toward “The use of computer-supported, interactive, visual TRENTON/ NEW BRUNSWICK. � representations of abstract data to simplify cognition.” (Card, NJ-18 N becomes CR-609 N/ METLARS LN. Mackinlay, Shneiderman, 1999) � Turn LEFT onto SUTPHEN RD. � Turn RIGHT onto FRELINGHUYSEN RD. 3 4
Information Visualization Goals of Information Visualization � Problem � Make large datasets coherent � How to understand massive datasets? � Present huge amounts of information compactly � Solution � Induce the viewer to think about the substance instead � Convert information into a graphical representation of methodology, design, technology, and so on � Take better advantage of human perceptual system � Encourage comparisons of different data � Issues � Present information at several levels of detail, from � What is a good visualization? overviews to fine structure � How to convert data? � Tell stories about the data statistically 5 6 Anscombe’s Scatter Plots Statistical Visualization: Anscombe's Quartet Data Set I Data Set II Data Set III Data Set IV Complex Positive X Y X Y X Y X Y non-linear linear 10.0 8.04 10.0 9.14 10.0 7.46 8.0 6.58 8.0 6.95 8.0 8.14 8.0 6.77 8.0 5.76 13.0 7.58 13.0 8.74 13.0 12.74 8.0 7.71 9.0 8.81 9.0 8.77 9.0 7.11 8.0 8.84 11.0 8.33 11.0 9.26 11.0 7.81 8.0 8.47 14.0 9.96 14.0 8.10 14.0 8.84 8.0 7.04 6.0 7.24 6.0 6.13 6.0 6.08 8.0 5.25 No variability 4.0 4.26 4.0 3.10 4.0 5.39 19.0 12.50 Linear w/ except 1 12.0 10.84 12.0 9.13 12.0 8.15 8.0 5.56 1 outlier 7.0 4.82 7.0 7.26 7.0 6.42 8.0 7.91 5.0 5.68 5.0 4.74 5.0 5.73 8.0 6.89 F.J. Anscombe, “Graphs in Statistical Analysis,” American Statistician , 27 (Feb 1973), pp17-21 7 8
Cholera Outbreak in London in 1854 John Snow’s Map of Cholera Deaths � The first death caused by cholera was found in London in 1831. � The year 1853 saw outbreaks in Newcastle and Gateshead as well as in London, where a total of 10,675 people died of the disease. � On August 31 of 1854, the outbreak of cholera hit London Soho area: 127 people died in the next three days and 500 within 10 days. � What is causing a cholera epidemic in London in 1854? � Dr. John Snow suspected cholera was transmitted by water, but could not prove it, Dr. John Snow plotted the location of then he used a map … deaths from cholera in central London for Sept 1854. Deaths are marked by black dots. Water pumps are marked with red cycles. 9 10 Time Series: Wheat Prices, Wages and Kings Today’s Time Series and Queens (William Playfair, 1786) King or Weekly wages of Price of quarter Queen good mechanics of wheat 11 12
Space & Time: Napoleon’s Army in Russia A More Readable Version (Charles Joseph Minard, 1861) “It may well be the best statistical graphic ever drawn.” Edward R. Tufte, 1983 13 14 Principles of Graphical Excellence Integrity Principle I � Graphical excellence is the well-designed presentation � The representation of numbers, as physically measured on the of interesting data – a matter of substance, of statistics, surface of the graphic itself, should be directly proportional to the and of design numerical quantities represented � Measure of violation � Graphical excellence consists of complex ideas communicated with clarity, precision, and efficiency Size of effect shown in graphic Lie Factor (LF) = � Graphical excellence is that which gives to the viewer Size of effect in data the greatest number of ideas in the shortest time with the least ink in the smallest space � Use logarithm of the Lie Factor to compare � Graphical excellence is nearly always multivariate � Overstating log LF > 0 � Graphical excellence requires telling the truth about the � Understating log LF < 0 data � Most distortions involve overstating; LF = 2-5 are common E.R. Tufte 1983 15 16
Example Violated the Principle Integrity Principle II � Clear, detailed, and thorough labeling Required Fuel Economy Standard: Fuel Economy Standards for Autos 1978 18 should be used to New Cars Built from 1978 to 1985 ‘79 20 defeat graphical Set by Congress and supplemented by the Transportation Department. ‘80 In miles per gallon. 22 30 27.7 distortion and ‘81 27 26 ambiguity 24 25 22 24 ‘82 � Write out explanations 20 Miles per Gallon 19 18 19.1 mpg expected 20 of the data on the average for all cars graphic itself 15 on road, 1985 13.7 mpg average 26 � Label important events for all cars on ‘83 10 in the data road, 1978 27 5 ‘84 27 1/2 0 ‘85 1978 1979 1980 1981 1982 1983 1984 1985 Adapted from The New York Times, August 9, 1978, p. D-2 17 18 Integrity Principle III Integrity Principle IV � Show data variation, not design variation � In time-series displays of money, deflated and standardized units of monetary measurement are always better than nominal units 19 20
Integrity Principle V LF = 1 � The number of � Now the area size of 1958 - Eisenhower: $1.00 1958 - Eisenhower: $1.00 information-carrying the dollar shrinks at the (variable) dimensions same rate as the dollar depicted should not value exceed the number of 1963 - Kennedy: 94¢ 1963 - Kennedy: 94¢ dimensions in the data 1968 - Johnson: 83¢ Dollar value shrinks in 1968 - Johnson: 83¢ one dimension, but the Purchasing Purchasing dollar sizes shrinks in 2 Power of the Power of the 1973 - Nixon: 64¢ dimensions 1973 - Nixon: 64¢ Diminishing Diminishing Dollar Dollar Adapted from The Washington Post, 1978 - Carter: 44¢ Source: Labor Department Source: Labor Department 1978 - Carter: 44¢ October 25, 1978, p.1 21 22 Integrity Principle VI Data Ink Principle (Tufte, 1983) � Graphics must not quote data out of context Down HW Data-ink 12% Data-ink 19% Recover Connecticut traffic deaths, = ratio 13% Total ink used to 1951-1959 325 SW 325 print the graphic Before stricter 15% 315 320 enforcement Labor 305 315 � Maximize the data-ink ratio 41% 295 310 Traffic deaths � Erase non-data-ink Traffic deaths 285 Down HW 305 � Erase redundant data-ink 19% 12% 275 300 Recover 265 13% 295 SW 255 290 15% 245 285 After stricter 235 280 enforcement Labor 225 275 41% 1 2 3 4 5 6 7 8 9 1955 1956 5 5 5 5 5 5 5 5 5 9 9 9 9 9 9 9 9 9 1 1 1 1 1 1 1 1 1 23 24
Example: Erase Redundant Data-Ink Data Maps: Cancer Mortality by County � What do we learn from the maps? 18 19 11 12 � What’s wrong with the data maps? 15 13 HW SW 16 Labor 15 Recover Down 46 43 1998 2000 11 Down 12 15 Recover 13 2000 46 Labor 43 1998 16 SW 15 18 HW 19 0 20 40 60 25 26 Data Maps: Cancer Mortality by SEA Data Maps: Cancer Mortality by SEA � What do you think about these maps? � What do we learn from these? 27 28
Data Maps: Breast Cancer by SEA Data Maps: Breast Cancer by SEA Black vs. White Female � Big difference between male and female? � What can 29 30 What About This Familiar Data Map Dimensions to Explore Data (Keim 97) Data Visualization Geometric Distortion Icon-based Pixel-oriented Hierarchical Complex Graph-based Simple Interaction Mapping Projection Filtering Link&Brush Zooming 31 32
Recommend
More recommend
