This document summarizes Krist Wongsuphasawat's work in data visualization at Twitter. It describes how he obtains tweet data, visualizes it using tools like R and D3 to show trends over time, locations, and text. Examples include visualizations of events like the World Cup and State of the Union. The process involves getting relevant data, visualizing it, evaluating the results, and iterating. The goal is to transform big Twitter data into smaller, insightful visualizations that tell stories.

1. visualization
at Twitter
data
Krist Wongsuphasawat / @kristw

2. Krist Wongsuphasawat / @kristw

3. Bangkok, Thailand
Krist Wongsuphasawat / @kristw

4. Computer Engineer
Bangkok, Thailand
Chulalongkorn University
Krist Wongsuphasawat / @kristw

5. Computer Engineer
Bangkok, Thailand
Programming + Soccer
Krist Wongsuphasawat / @kristw

6. Computer Engineer
Bangkok, Thailand
Programming + Soccer
Krist Wongsuphasawat / @kristw

7. Computer Engineer
Bangkok, Thailand
Programming + Soccer
Krist Wongsuphasawat / @kristw

8. Computer Engineer
Bangkok, Thailand
M.S. in Computer Science
Univ. of Maryland
Krist Wongsuphasawat / @kristw

9. Computer Engineer
Bangkok, Thailand
PhD in Computer Science
Univ. of Maryland
Information Visualization
Krist Wongsuphasawat / @kristw

10. Computer Engineer
Bangkok, Thailand
PhD in Computer Science
Univ. of Maryland
Information Visualization
Krist Wongsuphasawat / @kristw
IBM
Microsoft

11. Computer Engineer
Bangkok, Thailand
PhD in Computer Science
Univ. of Maryland
Information Visualization
Krist Wongsuphasawat / @kristw
IBM
Microsoft
Sr. Data Visualization Scientist
Twitter

12. data visualization
at Twitter

13. data visualization
at Twitter

14. visualization
data
at Twitter

15. vis
data
at Twitter

16. data
at Twitter
“Tweets”

17. data
at Twitter
“Tweets”
#events
TV Shows New Year
Earthquake
Oscars
Protest
Super Bowl
World Cup Election
Breaking news
…

18. data
at Twitter
“Tweets”
#events
TV Shows New Year
Earthquake
Oscars
Protest
Super Bowl
World Cup Election
Breaking news
…
#curiosity
Sleep pattern
Human behavior
Language …

19. data
at Twitter
“Tweets”
#events
TV Shows New Year
Earthquake
Oscars
Protest
Super Bowl
World Cup Election
Breaking news
…
#curiosity
Sleep pattern
Human behavior
Language …
What could we learn from the Tweets?

20. vis
data
at Twitter
“Tweets”
Tell stories about an event,
Pursue curiosity or inspiration
Goal:

21. vis
data
at Twitter
“Tweets”
Tell stories about an event,
Pursue curiosity or inspiration
(with deadline)
Goal:

22. Challenge accepted

23. vis
data
at Twitter
“Tweets”
Get data
1

24. easy?

25. Having all Tweets
How people think I feel.

26. Having all Tweets
How people think I feel. How I really feel.

27. Challenges
• Too much data
• Want only relevant Tweets
• hashtag: #BRA
• keywords: “goal”
• Need to aggregate & reduce size
• Long processing time (hours)

28. Hadoop Cluster
Vertica
Pig / Scalding (slow) SQL
Data Storage
Tool
Workflow

29. Hadoop Cluster
Vertica
Pig / Scalding (slow) SQL
Data Storage
Tool
Workflow

30. Workflow
Hadoop Cluster
Vertica
Pig / Scalding (slow) SQL
Data Storage
Tool
Your laptop Smaller dataset

31. Hadoop Cluster
Vertica
Pig / Scalding (slow) SQL
Data Storage
Tool
Tool node.js / python / excel (fast)
Final dataset
Your laptop
Workflow
Smaller dataset

32. vis
data
at Twitter
“Tweets”
Get data
1
2
Visualize

33. Visualize
• Peek into data
• Check data & test ideas
• Decide how to visualize
• Guided by data type
• Choose tools
• Start building

34. Visualize
• Peek into data
• Check data & test ideas
• Decide how to visualize
• Guided by data type
• Choose tools
• Start building
R d3
Tableau Yeoman

35. (+ media)
photos, videos
data
What?
TEXT
Where? When?
GEO TIME

36. Visualize Data
What?
TEXT
Where? When?
GEO TIME

37. Visualize Data
What?
TEXT
Where? When?
GEO TIME

38. Time Tweets/second

39. Time Tweets/second

40. Time Tweets/second + Annotation
http://www.flickr.com/photos/twitteroffice/5681263084/

41. Visualize Data
What?
TEXT
Where? When?
GEO TIME

42. Geo
Heatmap
Low density
High density

43. Geo
San Francisco
flickr.com/photos/twitteroffice/8798020541
Low density
High density

44. Geo
San Francisco
Rebuild the world
based on
tweet volumes
twitter.github.io/interactive/andes/

45. Visualize Data
What?
TEXT
Where? When?
GEO TIME

46. Text
www.wordle.net
Some experiments
during World Cup

47. Text
www.wordle.net
Word cloud of Tweets right after the 1st goal

48. Text Word cloud of Tweets right after the 1st goal
It was an “own” goal.
www.wordle.net

49. Text WordTree [Wattenberg & Viégas 2008]
www.jasondavies.com/wordtree
www.jasondavies.com/wordtree

50. Visualize Data
What?
TEXT
Where? When?
GEO TIME

51. Time + Geo
Japan Earthquake 2011
blog.twitter.com/2011/global-pulse
youtu.be/SybWjN9pKQk

54. Time + Geo Tweet pattern [Rios & Lin 2012]
Night
Late night
Daytime
Night
Late night
Daytime

55. Time + Geo Tweet pattern [Rios & Lin 2012]
Night
Late night
Daytime
Night
Late night
Daytime

56. Time + Geo Tweet pattern [Rios & Lin 2012]
Night
Late night
Daytime
Night
Late night
Daytime

57. Time + Geo Tweet pattern [Rios & Lin 2012]
Night
Late night
Daytime
Night
Late night
Daytime

58. Visualize Data
What?
TEXT
Where? When?
GEO TIME

59. Geo + Text Real-time Tweet map

60. Geo + Text Real-time Tweet map

61. Geo + Text Real-time Tweet map
most
frequent
term

62. Geo + Text Real-time Tweet map
Gmail was down
Jan 24, 2014

63. Geo + Text Real-time Tweet map
Nelson Mandela
passed away
Dec 5, 2013

64. Visualize Data
What?
TEXT
Where? When?
GEO TIME

65. Time + Text
UEFA Champions League
Biggest tournament for European soccer clubs
Many Tweets during the matches

66. UEFA Champions League
Team 1 Team 2
Time + Text
Dortmund Bayern Munich

67. UEFA Champions League
Team 1 Team 2
Time + Text
Dortmund Bayern Munich

68. UEFA Champions League
Team 1 Team 2
Time + Text
Dortmund Bayern Munich

69. UEFA Champions League
Team 1 Team 2
Dortmund Bayern Munich
Count Tweets mentioning
the teams every minute
Time + Text

70. Time + Text UEFA Champions League

71. Time + Text UEFA Champions League
+ “goal” count
+ context

72. + “offside”
Time + Text UEFA Champions League

73. + players
Time + Text UEFA Champions League

74. Competition Tree
vs vs
A B C D
vs
A C
C

75. Competition Tree
vs vs
A B C D
vs +
A C
C

76. Competition Tree
vs vs
A B C D
vs + =
A C
C

77. Visualize Data
What?
TEXT
Where? When?
GEO TIME

78. Time + Text + Geo State of the Union
twitter.github.io/interactive/sotu2014

79. Time + Text + Geo State of the Union
1) timeline + topic from Tweets
4) Density map of
Tweets about
selected topic
3) Volume of Tweets
by topics
during selected
part of the SOTU
2) context
(speech)
twitter.github.io/interactive/sotu2014

80. Time + Text World Cup 2014

81. Time + Text + Geo World Cup 2014

82. Visualize Data
What?
TEXT
Where? When?
GEO TIME

83. Visualize Data
What?
TEXT
+
Where? When?
GEO TIME
Non-Twitter data
CONTEXT

84. Time + Text New Year 2014

85. Time + Text New Year 2014

86. Time + Text + Geo (c) New Year 2014
twitter.github.io/interactive/newyear2014/

87. vis
data
at Twitter
“Tweets”
Get data
1
2
Visualize

88. vis
data
at Twitter
“Tweets”
Get data
1
2
Visualize
Evaluate
3

89. vis
data
at Twitter
“Tweets”
Get data
1
2
Visualize
Evaluate
3
Iterate!

90. Evaluation
• Self
• Peer feedback
• Non team members / Potential audience

91. vis
data
at Twitter
Get data
1
2
Visualize
Evaluate
3

92. vis
data
at Twitter
Get data
1
2
Visualize
Evaluate
3
big data => small data

93. vis
data
at Twitter
Get data
1
2
Visualize
Evaluate
3
big data => small data
What? Where? When?

94. big data => small data self, peer, external
vis
data
at Twitter
Get data
1
2
Visualize
Evaluate
3
What? Where? When?

95. big data => small data self, peer, external
vis
data
at Twitter
“Tweets”
Get data
1
2
Visualize
Evaluate
3
What? Where? When?

96. big data => small data self, peer, external
vis
data
at Twitter
“Tweets”
Get data
1
2
Visualize
Evaluate
3
What? Where? When?
• users
• followers graph
• logs
• etc.
!
• derived data: language, sentiment

97. big data => small data self, peer, external
vis
data
at Twitter
“Tweets”
Get data
1
2
Visualize
Evaluate
3
What? Where? When?
• users Who? …
• followers graph
• logs
• etc.
!
• derived data: language, sentiment

98. big data => small data self, peer, external
vis
data
at Twitter
“Tweets”
Get data
1
2
Visualize
Evaluate
3
What? Where? When?
• users Who? …
• followers graph
• logs
• etc.
!
• derived data: language, sentiment
(with deadline)

99. big data => small data self, peer, external
vis
data
at Twitter
“Tweets”
Get data
1
2
Visualize
Evaluate
3
What? Where? When?
• users Who? …
• followers graph
• logs
• etc.
(with deadline)
!
• derived data: language, sentiment @kristw / https://interactive.twitter.com

100. big data => small data self, peer, external
vis
data
at Twitter
“Tweets”
Get data
1
2
Visualize
Evaluate
3
What? Where? When?
• users Who? …
• followers graph
• logs
• etc.
(with deadline)
!
• derived data: language, sentiment @kristw / https://interactive.twitter.com
+ visualizations by @philogb, @miguelrios & @trebor

101. Questions?

102. big data => small data self, peer, external
vis
data
at Twitter
“Tweets”
Get data
1
2
Visualize
Evaluate
3
What? Where? When?
• users Who? …
• followers graph
• logs
• etc.
(with deadline)
@kristw / https://interactive.twitter.com
+ visualizations by @philogb, @miguelrios & @trebor

103. Thank you