Hacking Data Visualisations

A look at data visualisations though the ages, and an introduction to R

Transcript

1. Hacking Data Visualisations MELINDA SECKINGTON ! @MSECKINGTON
2. @mseckington
6. Hacking data visualisations @mseckington
7. Why?
8. https://www.flickr.com/photos/laurenmanning/6632168961/
9. https://www.flickr.com/photos/jamjar/5491205608
10. “I feel that everyday, all of us now are being blasted by information design. It's being poured into our eyes through the Web, and we're all visualizers now; we're all demanding a visual aspect to our information. There's something almost quite magical about visual information. It's effortless, it literally pours in. And if you're navigating a dense information jungle, coming across a beautiful graphic or a lovely data visualization, it's a relief, it's like coming across a clearing in the jungle.” DAVID MCCANDLESS - THE BEAUTY OF DATA VISUALIZATION @mseckington
11. Tor Norretranders THE BANDWIDTH OF OUR SENSES @mseckington
16. A brief history of data visualisations
17. Theatrum Orbis Terrarum May 20, 1570 The first modern atlas, collected by Abraham Ortelis. ! This was a first attempt to gather all maps that were known to man at the time and bind them together. A BRIEF HISTORY OF DATA VISUALISATION
18. https://www.flickr.com/photos/smailtronic/2361594300
19. A BRIEF HISTORY OF DATA VISUALISATION Bills of Mortality From 1603, London parish clerks collected health- related population data in order to monitor plague deaths, publishing the London Bills of Mortality on a weekly basis. ! John Graunt amalgamated 50 years of information from the bills, producing the first known tables of public health data. BEAUTIFUL SCIENCE AT THE BRITISH LIBRARY - THE GUARDIAN
20. A BRIEF HISTORY OF DATA VISUALISATION 1644: First known graph of statistical data ! MICHAEL VAN LANGREN - ESTIMATES OF DISTANCE IN LONGITUDE BETWEEN TOLEDO AND ROME
21. A BRIEF HISTORY OF DATA VISUALISATION
22. A BRIEF HISTORY OF DATA VISUALISATION 1786 first bar chart William Playfair Exports and imports of Scotland to and from different parts for one Year from Christmas 1780 to Christmas 1781
23. A BRIEF HISTORY OF DATA VISUALISATION Street map of cholera deaths in Soho 1853 John Snow Snow's 'ghost map' shows deaths from cholera around Broad Street between 19 August and 30 September 1854. Snow simplified the street layout, highlighting the 13 water pumps serving the area and representing each death as a black bar. His map demonstrates how cholera was spreading, not by a 'miasma' rising from the Thames, but in water contaminated by human waste BEAUTIFUL SCIENCE AT THE BRITISH LIBRARY - THE GUARDIAN
24. A BRIEF HISTORY OF DATA VISUALISATION Diagram of the Causes of Mortality in the Army in the East ! 1858 Florence Nightingale In her seminal ‘rose diagram’, Nightingale demonstrated that far more soldiers died from preventable epidemic diseases (blue) than from wounds inflicted on the battlefield (red) or other causes (black) during the Crimean War (1853-56) BEAUTIFUL SCIENCE AT THE BRITISH LIBRARY - THE GUARDIAN
25. How?
26. HOW? https://www.flickr.com/photos/jdhancock/8031897271
27. https://www.flickr.com/photos/laurenmanning/5658951917/
28. HOW? @mseckington
29. HOW? @mseckington
30. HOW? @mseckington
31. HOW? @mseckington
32. HOW? @mseckington
33. A quick intro to R
34. A QUICK INTRO TO R What is R? ! @mseckington
35. A QUICK INTRO TO R What is R? ! R is a free programming language and environment for statistical computing and graphics. ! @mseckington
36. A QUICK INTRO TO R What is R? ! R is a free programming language and environment for statistical computing and graphics. ! Created by statisticians for statisticians. @mseckington
37. A QUICK INTRO TO R What is R? ! R is a free programming language and environment for statistical computing and graphics. ! Created by statisticians for statisticians. ! Comes with a lot of facilities for data manipulation, calculation, data analysis and graphical display. @mseckington
38. A QUICK INTRO TO R What is R? ! R is a free programming language and environment for statistical computing and graphics. ! Created by statisticians for statisticians. ! Comes with a lot of facilities for data manipulation, calculation, data analysis and graphical display. ! Highly and easily extensible. @mseckington
39. A QUICK INTRO TO R
41. ! > data()! ! list all datasets available ! @mseckington
42. ! > data()! ! list all datasets available ! > movies = data(movies)! > movies <- data(movies)! ! assign movies data to movies variable ! @mseckington
43. ! > data()! ! list all datasets available ! > movies = data(movies)! > movies <- data(movies)! ! assign movies data to movies variable ! > dim(movies)! [1] 58788! 24! ! @mseckington
44. ! > data()! ! list all datasets available ! > movies = data(movies)! > movies <- data(movies)! ! assign movies data to movies variable ! > dim(movies)! [1] 58788! 24! ! > names(movies)! [1] "title" “year" “length" “budget" "rating" “votes" ! [7] “r1" “r2" “r3" “r4" “r5" “r6"! [13] “r7" “r8" “r9" “r10" “mpaa" “Action" ! [19] “Animation" "Comedy" “Drama" “Documentary" “Romance”"Short"! @mseckington
45. ! > movies[7079,]! ! !! title ! ! ! ! ! year ! length budget rating votes ! 7079 Bourne Identity, The 2002 ! 119! ! 75000000 7.3 ! 29871 ! ! r1 r2 r3 r4 r5 r6 r7 r8 r9 r10 mpaa ! 4.5 4.5 4.5 4.5 4.5 14.5 24.5 34.5 14.5 4.5 PG-13! ! Action Animation Comedy Drama Documentary Romance Short! 1 0 0 1 0 0 0! ! returns 1 row => all the data for 1 movies ! @mseckington
46. ! > movies[7079,]! ! !! title ! ! ! ! ! year ! length budget rating votes ! 7079 Bourne Identity, The 2002 ! 119! ! 75000000 7.3 ! 29871 ! ! r1 r2 r3 r4 r5 r6 r7 r8 r9 r10 mpaa ! 4.5 4.5 4.5 4.5 4.5 14.5 24.5 34.5 14.5 4.5 PG-13! ! Action Animation Comedy Drama Documentary Romance Short! 1 0 0 1 0 0 0! ! returns 1 row => all the data for 1 movies ! > movies[1:10,]! . . . ! ! returns rows 1 to 10 @mseckington
47. ! > movies[,1]! . . .! ! returns 1 column => titles of all movies @mseckington
48. ! > movies[,1]! . . .! ! returns 1 column => titles of all movies ! > movies$title! . . .! ! same as movies[,1]! returns column with the label ‘title ! @mseckington
49. ! > movies[,1]! . . .! ! returns 1 column => titles of all movies ! > movies$title! . . .! ! same as movies[,1]! returns column with the label ‘title ! > movies[,1:10]! . . .! ! returns columns 1 to 10 @mseckington
50. ! > hist(movies$year) @mseckington
51. ! > hist(movies$year) Histogram of movies$year movies$year Frequency 1900 1920 1940 1960 1980 2000 0 2000 4000 6000 8000 @mseckington
52. ! > hist(movies$year)! ! > hist(movies$rating) @mseckington
53. ! > hist(movies$year)! ! > hist(movies$rating) Histogram of movies$rating movies$rating Frequency 2 4 6 8 10 0 2000 4000 6000 8000 @mseckington
54. ! > hist(movies$year)! ! > hist(movies$rating)! ! > library(ggplot2) @mseckington
55. ! > hist(movies$year)! ! > hist(movies$rating)! ! > library(ggplot2)! ! > qplot(rating, ! ! ! !! data=movies, ! !! geom="histogram") @mseckington
56. ! > hist(movies$year)! ! > hist(movies$rating)! ! > library(ggplot2)! ! > qplot(rating, ! ! ! !! data=movies, ! !! geom=“histogram")! ! > qplot(rating, ! ! !! data=movies, ! !! geom="histogram", !! binwidth=1) @mseckington
57. ! > m = ggplot(movies, aes(rating))! ! > m + geom_histogram() @mseckington
58. ! > m = ggplot(movies, aes(rating))! ! > m + geom_histogram()! ! > m + geom_histogram(! ! ! ! aes(fill = ..count..)) @mseckington
59. ! > m = ggplot(movies, aes(rating))! ! > m + geom_histogram()! ! > m + geom_histogram(! ! ! ! aes(fill = ..count..))! ! > m + geom_histogram(! ! ! ! colour = "darkgreen", ! ! ! ! fill = "white", ! ! ! ! binwidth = 0.5)! ! @mseckington
60. ! > m = ggplot(movies, aes(rating))! ! > m + geom_histogram()! ! > m + geom_histogram(! ! ! ! aes(fill = ..count..))! ! > m + geom_histogram(! ! ! ! colour = "darkgreen", ! ! ! ! fill = "white", ! ! ! ! binwidth = 0.5)! ! > x = m + geom_histogram(! ! ! ! ! binwidth = 0.5)! > x + facet_grid(Action ~ Comedy)! @mseckington
61. ! > library(twitteR)! ! > setup_twitter_oauth(! ! ! "API key”, "API secret", "Access token", "Access secret”)! ! @mseckington
62. FUTURELEARN STATS
63. ! > fl = read.csv(! ! ! "futurelearn_dataset.csv", ! ! header=TRUE)! ! @mseckington
64. ! > fl = read.csv(! ! ! "futurelearn_dataset.csv", ! ! header=TRUE)! ! > source_table = table(fl$age)! > pie(source_table) @mseckington
65. ! > fl = read.csv(! ! ! "futurelearn_dataset.csv", ! ! header=TRUE)! ! > source_table = table(fl$age)! > pie(source_table)! ! > pie(source_table, ! ! ! radius=0.6, ! ! ! col=rainbow(8)) @mseckington
67. ! > library(twitteR)! ! > setup_twitter_oauth(! ! ! "API key”, "API secret", "Access token", "Access secret”)! ! > tweets <- searchTwitter('futurelearn', n=100) @mseckington
69. ! > library(twitteR)! ! > setup_twitter_oauth(! ! ! "API key”, "API secret", "Access token", "Access secret”)! ! > tweets <- searchTwitter('futurelearn', n=100)! ! > library(“tm”)! ! > tweet_text <- sapply(tweets, function(x) x$getText())! > tweet_corpus <- Corpus(VectorSource(tweet_text))! ! @mseckington
70. ! > library(twitteR)! ! > setup_twitter_oauth(! ! ! "API key”, "API secret", "Access token", "Access secret”)! ! > tweets <- searchTwitter('futurelearn', n=100)! ! > library(“tm”)! ! > tweet_text <- sapply(tweets, function(x) x$getText())! > tweet_corpus <- Corpus(VectorSource(tweet_text))! ! > tweet_corpus <- tm_map(tweet_corpus, ! ! ! ! ! ! ! ! ! ! ! content_transformer(tolower))! > tweet_corpus <- tm_map(tweet_corpus, removePunctuation)! > tweet_corpus <- tm_map(tweet_corpus, ! ! ! ! ! ! ! ! ! ! ! function(x)removeWords(x,stopwords()))
72. ! > library(wordcloud)! ! > wordcloud(tweet_corpus) @mseckington
73. ! > library(wordcloud)! ! > wordcloud(tweet_corpus) @mseckington
74. What next?
75. A QUICK INTRO TO R
76. A QUICK INTRO TO R
78. WHAT NEXT? @mseckington
79. https://www.flickr.com/photos/jamjar/5491205608
80. @mseckington
81. Recap
82. Data visualisations are awesome @mseckington
83. R is awesome @mseckington
84. Any questions? ! @mseckington