Design of Charts Prof. Michael McGuffin

1. Design of ChartsProf. Michael McGuffin

2. Examples of relations. A relationis a subset of a cartesian product of two or more sets. Each relation can be thought of as a multidimensional data set. In the examples here, each row is a tuple, each column is a dimension. The fonction y = x^0.5: x y --- --- 0 0 1 1 4 2 9 3 ... The relation in a table of a relational database: Client_name Product_purchased Price Date ... ------------- ----------------- ------- ------------ ----- Robert G. Trombone 500.00 2008 mars 7 . Robert G. Partitions vol. 1 45.00 2008 mars 7 . Lucie M. Flute 180.00 2007 nov 11 . Cynthia S. Partitions vol. 2 40.00 2008 juin 16 Jules T. Piano 6000.00 2008 jan 10 Jules T. Partitions vol. 1 45.00 2008 jan 13 ... A video (for example, an .avi file): x y time red green blue --- --- ------- ------- ------ ------ 0 0 0 255 0 0 0 1 0 200 10 6 ... 0 0 0.1 255 50 100 0 1 0.1 255 200 190 ...

3. Tableau: software for visualizing databases(Mackinlay et al. 2007, tableausoftware.com)

9. Columns: a, x Rows: b, y b y y x x y y x x a

11. Tableau For more information:http://www.tableausoftware.com/products/tourhttp://www.tableausoftware.com/products/desktop/demo

12. Types of dimensions • Quantitative (or continuous, or metric) • Examples: x, y, time, temperature • Ordinal • We can put the values in a logical order, but we cannot say that one value is N times bigger than another value • Example: high school diploma, bachelor’s degree, master’s degree,doctoral degree (in order of years of schooling) • Nominal (or categorical) • There is no natural order (except maybe alphabetical, which is arbitrary and language-dependent) • Example: degree in mechanical engineering, degree in physics, degree in math, degree in music, etc. • Example: Honda, Toyota, GM, Chrysler, etc. • Binary • A kind of nominal (or ordinal?) dimension with two possible values

13. Output graphical representation:at most 3D space × 1D time Input data:dimensions can be {independent, dependent}and{continuous, ordinal, nominal}

14. Hierarchy of graphical variables

15. Example from a course by Marilyn Ostergren at U Washington( http://courses.washington.edu/info424/Week3Practice_ExcelGraphs.html )

16. Hierarchy of graphical variables (Mackinlay, 1986)

17. Tableau • Automatically determines which columns in a database are "dimensions" (independent variables), which are "measures" (dependent variables), and which are "quantitative" (continuous) or "categorical" (nominal). • Automatically chooses a kind of chart or graphic based on the nature of the data (Mackinlay et al. 2007)

18. Tableau Examples resulting from applying the rules on the previous slide:

21. Bar chart vs line graph Which makes it easier to perceive changes in slope?

22. Length vs area Tufte (1983)

25. From IEEE Canadian Review, 2009, No. 60, page 31

26. Example from a course by Marilyn Ostergren at U Washington( http://courses.washington.edu/info424/Week3Practice_ExcelGraphs.html )

28. http://www.research.ibm.com/people/l/lloydt/color/color.HTMRogowitz and Treinish, “Why Should Engineers and Scientists Be Worried About Color?”

29. Borland and Taylor, “Rainbow Color Map (Still) Considered Harmful”, IEEE CG&A, 27(2):14-17, 2007

32. ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000 ABC abc 123 000

33. Other examples … Rows are located on grey bands as well as between grey bands. The number of grey bands necessary is half the number of separating lines that would be necessary between rows. Notes are shown on lines as well as between lines, reducing the number of lines necessary by a factor of 2.

34. Naïve notation: semitone octave octave Modern notation: octave

35. Genetic code (mapping fromnucleotide triplets to amino acids) Ben Fry’s version (http://benfry.com/aasd/) Traditional versions

36. Graphic invented by Florence Nightingale(1820-1910; statistician, and pioneer in nursing care) http://upload.wikimedia.org/wikipedia/commons/1/17/Nightingale-mortality.jpg

37. http://www.economist.com/images/20071222/5107CR3B.jpg

38. A polar variant of parallel coordinates Noms: star plots, star glyphs, star coordinates, spider chart, radar chart, polar chart, kiviat diagram. http://en.wikipedia.org/wiki/Radar_chart

39. A polar variant of parallel coordinates Stephen Few; http://www.perceptualedge.com/example4.php

40. A polar variant of parallel coordinates http://www.onscale.de/specbrowser/

41. Class exercise:Design one or more graphs to visualize a data setof 19 tuples, with the following dimensions: • Car model: {Accord, AMC Pacer, Audi 5000, BMW 320i, Champ, Chev Nova, …} (19 models in total) • Car price: [$0, $13500] • Car mileage: [0,40] • Repair record: {Great, Good, OK, Bad, Terrible} • Car weight: [0,5500] Most important dimensions

43. Exercise en classe:Concevoir un ou des graphiques pour visualiser un jeu de données de 19 points, ayant les dimensions suivantes: • Modèle d’auto: {Accord, AMC Pacer, Audi 5000, BMW 320i, Champ, Chev Nova, …}(19 modèles en tout) • Prix d’auto: [$0, $13500] • Consommation: [0,40] • Historique de réparations: {Excellent, Bon, OK, Mauvais, Affreux} • Poids: [0,5500] Dimensions les plus importantes