Information Visualization Crash Course

Transcription

1 Information Visualization Crash Course (AKA Information Visualization 101) Chad Stolper Assistant Professor Southwestern University (graduated from Georgia Tech CS PhD) 1

2 What is Infovis? Why is it Important? Human Perception Chart Basics (If Time, Some Color Theory) The Shneiderman Mantra Where to Learn More 2

3 What is Information Visualization? 3

4 Information Visualization The use of computer-supported, interactive, visual representations of abstract data to amplify cognition. Card, Mackinlay, and Shneiderman

5 Communication Exploratory Data Analysis (EDA) 5

6 Communication (gone wrong) 6

7 7

8 Edward Tufte An American statistician and professor emeritus of political science, statistics, and computer science at Yale University. He is noted for his writings on information design and as a pioneer in the field of data visualization. -Wikipedia 8 X

9 Space Shuttle Challenger January 28, 1986 Morning Temperature: 31 F 9

10 10

11 Tufte, E. R. (2012). Visual explanations: images and quantities, evidence and narrative. Cheshire, CT: Graphics Press. 11

12 Most Watched Science Experiment Richard Feynman, Physics Nobel laureate explained how rubber became rigid in cold temperate YouTube video: Video originally from: 13

13 How did this happen? 14

14 Engineers at Morton Thiokol, the rocket maker, presented on the day before and recommended not to launch. Tufte, E. R. (2012). Visual explanations: images and quantities, evidence and narrative. Cheshire, CT: Graphics Press. 15

15 19

16 24

17 25

18 26

19 27

20 28

21 So, communication is extremely important. Visualization can help with that communicate ideas and insights. 29

22 30

23 Visualization can also help with Exploratory Data Analysis (EDA) But why do you need to explore data at all??? 31

24 There are three kinds of lies: lies, damned lies, and statistics. 33

25 Mystery Data Set 34

26 Mystery Data Set Property Value mean( x ) 9 variance ( x ) 11 mean( y ) 7.5 variance ( y ) correlation ( x,y ) Linear Regression Line y = x 35

27 36

28 37

29 38

30 39

31 Anscombe s Quartet 40

32 Anscombe s Quartet Sanity Checking Models Outlier Detection 41

33 Data visualization leverages human perception 43

34 Name the five senses. 44

35 Sense Bandwidth (bits/sec) Sight 10,000,000 Touch 1,000,000 Hearing 100,000 Smell 100,000 Taste 1,

36 A (Simple) Model of Human Visual Perception 46

37 A (Simple) Model of Human Perception Stage 1 Stage 2 Parallel detection of basic features into an iconic store Serial processing of object identification and spatial layout 47

38 Stage 1: Pre-Attentive Processing Rapid Parallel Automatic (Fleeting = lasting for a short time) 48

39 Stage 2: Serial Processing Relatively Slow (Incorporates Memory) Manual 49

40 Stage 1: Pre-Attentive Processing The eye moves every 200ms (so this processing occurs every 200ms-250ms) 50

41 Example

42 Example

43 A few more examples from Prof. Chris Healy at NC State 53

44 Left Side Right Side 54

45 Raise your hand if a RED DOT is present (On the left or on the right?) 55

46 56

47 57

48 Color (hue) is pre-attentively processed. 58

49 Raise your hand if a RED DOT is present 59

50 60

51 61

52 Shape is pre-attentively processed. 62

53 Determine if a RED DOT is present 63

54 64

55 65

56 Hue and shape together are NOT pre-attentively processed. 66

57 Pre-Attentive Processing length width size curvature number terminators intersection closure hue lightness flicker direction of motion binocular lustre stereoscopic depth 3-D depth cues lighting direction 67

58 Stephen Few Now You See It pg

59 Pre-Attentive à Cognitive 69

60 Gestalt Psychology Berlin, Early 1900s 70

61 Gestalt Psychology Goal was to understand pattern perception Gestalt (German) = seeing the whole picture all at once instead of a collection of parts Identified 8 Laws of Grouping 71

62 Gestalt Psychology 1. Proximity 2. Similarity 7. Good Gestalt 8. Past Experience 3. Closure 4. Symmetry 5. Common Fate 6. Continuity 72

63 How many groups are there? 73

64 74

65 Proximity 75

66 How many groups are there? 76

67 77

68 Similarity 78

69 How many shapes are there? 79

70 80

71 Closure 81

72 How many items are there? 82

73 ( ) { } [ ] 83

74 Symmetry ( ) { } [ ] 84

75 How many sets are there? 85

76 86

77 Common Fate 87

78 How many objects are there? 88

79 89

80 Continuity 90

81 How many objects are there? 91

82 92

83 Good Gestalt 93

84 What is this word? 94

85 FLIGHT 95

86 Past Experience FLIGHT 96

87 Pre-Attentive Processing Gestalt Laws 101

88 Detect Quickly 102

89 Detect quickly does NOT mean detect accurately Ideally you want both. 103

90 Angles Positions Circular areas Rectangular areas (aligned or in a treemap) Crowdsourcing Graphical Perception: Using Mechanical Turk to Assess Visualization Design.Heer and Bostock. Proc ACM Conf. Human Factors in Computing Systems (CHI) 2010, p

91 Crowdsourced Results Positions Angles Circular areas Rectangular areas (aligned or in a treemap) Log Error Crowdsourcing Graphical Perception: Using Mechanical Turk to Assess Visualization Design.Heer and Bostock. Proc ACM Conf. Human Factors in Computing Systems (CHI) 2010, p

92 More accurate I Position I IMll 1 I Length F-l Less accurate Iha I0.I I rl l kj cl Color mot Shown) Volume I I Mackinlay,

93 Stephen Few Now You See It pg

94 What does this tell us? 108

95 Barcharts, scatterplots, and line charts are really effective for quantitative data

96 (and for statistical distributions) Tukey Box Plots 113

97 Outliers Largest < Q IQR Largest < Q3 Median Smallest > Q1 Smallest > Q1-1.5 IQR 114

98 115

99 Tufte s Chart Principles Edward Tufte 116

100 Tufte s Chart Principles DO NOT LIE! Maximize Data-Ink Ratio Minimize Chart Junk 119

101 Tufte s Chart Principles DO NOT LIE! Maximize Data-Ink Ratio Minimize Chart Junk 120

102 121

103 Cumulative 122

104 123

105 124

106 Tufte s Chart Principles DO NOT LIE! Maximize Data-Ink Ratio Minimize Chart Junk 126

107 127

108 Chartjunk. (2017, October 05). Retrieved December 01, 2017, from 128

109 Please 130

110 No pie charts. No 2.5D charts. 131

111 132

112

113

114 135

115 PLEASE DON T EVER DO THIS! 136

116

117 But otherwise 141

118 Barcharts, scatterplots, and line charts are really effective for quantitative data

119 Anyone else bored by my color choices? 143

120 In fact, grayscale can be risky 144

121 In fact, grayscale can be risky 145

122 Color is Powerful 146

123 Color Call attention to information Increase appeal Increase memorability Another dimension to work with 147

124 Have you heard of RGB? Additive color model: colors create by mixing red, green, blue light RGB color model. (2017, November 20). Retrieved December 01, 2017, from 148

125 We see in RGB, but we don t interpret in RGB 149

126 HSV Color Model Hue Saturation Source: color picker in Affinity Designer Lightness 150

127 Hue Post & Greene,

128 Hue 152

129 Hue and Colorblindness 10% of males and 1% of females are Red-Green Colorblind 153

130 154

131 155

132 Color and Quantitative Data Can you order these (lowàhi)? 157

133 Binary Categorical Categorical Categorical Diverging Sequential via Munzner 158

134 Color Brewer for Picking Color Scales COLORBREWER 2.0. (n.d.). Retrieved December 01, 2017, from 159

135 Overview Zoom+Filter Details on Demand Shneiderman Mantra (Information-Seeking Mantra) 160

136 161

137 162

138 163

139 Where to learn more? 172

140 CS 7450 Information Visualization Every Fall 173

141 vis.gatech.edu 174

142 How to Make Good Charts Edward Tufte s One-Day Workshop Edward Tufte, Visual Display of Quantitative Information Stephen Few, Show Me the Numbers: Designing Tables and Graphs to Enlighten Designing- Enlighten/dp/ /ref=la_B001H6IQ5M_1_ 2?s=books&ie=UTF8&qid= &sr=

143 Visualization Theory Books Tamara Munzner VIS Tutorial and Book Colin Ware, Information Visualization: Perception for Design Technologies/dp/ Stephen Few, Now You See It Quantitative/dp/ /ref=pd_bxgy_b_img_z Edward Tufte, Envisioning Information Edward Tufte, Visual Explanations Edward Tufte, Beautiful Evidence Tamara Munzner, Visualization Analysis & Design Peters/dp/

144 Perception and Color Websites Chris Healy, NC State tml Color Brewer Maureen C. Stone (Color Links, Blog, Workshops) Subtleties of Color by Robert Simmon of NASA 177

145 Visualization Blogs Flowing Data by Nathan Yau Information Aesthetics by Andrew Vande Moere Information is Beautiful by David McCandless Visual.ly Blog Indexed Comic by Jessica Hagy 178

146 Infographics Visual.ly/view (wtfviz.net) 179

147 Thanks! Chad Stolper 180

148 Jessica Hagy thisisindexed.com Questions? Chad Stolper 181