an analysis of death spikes on celebrity wikipedia pages created for gslis 590mt informetrics

1. Death and edits
Miles Lincoln
LIS590MT

2. Wikipedia!
You probably already know
what it is!

3. Bursts in social networks
 Bursts of edits on Wikipedia in particular
 When do those occur?

4. What can we learn by looking at
spikes in edit frequency?
 How have edit spikes changed over Wikipedia’s ten
years of existence?
 Does the size of an edit spike correlate to anything?

5. Bursts in other social networks

6. Google Trends

7. Celebrity deaths!

8. Revision history

9. Revision history

10. But first…
 We need to process the data so that we can answer that
question

11. Perl

12. Regular Expressions (Regex)
Perl script uses regular expressions to find and
output matching pieces of text.
In this case, I am pulling out dates in Wikipedia’s
day month year format and re-writing them in a
more machine-readable MM/DD/YYYY format.
11/08/2011

13. Data manipulation
Copy/pase the revision history of wiki
pages into a text document which I
feed to my perl script
Results in lists consisting of one date
per edit that occurred on that date
Copying/pasting isn’t super
elegant, but I haven’t gotten
LWP/useragent stuff to work yet

14. Excel!
 Throw my lists of dates into a pivot table, which
shows me the frequency that each date occurs
 Some vlookup magic allows me to combine
these edit frequencies of individual actors into
one big list covering every day from 6/1/2001 to
the present

15. Et Voila!

16. Problems
9 actors over 10 years means close to 100k cells
Excel is not built for speed
Matlab might work better

17. What does the data look like over
time?
 6/1-5/31 from 2001 (when Wikipedia’s current edit no.’s
begin) to 2010 (when all of the bursts have settled down)

18. 6/1/2001-5/31/2002
1.2
1
0.8 Series1
Series2
Series3
Series4
0.6
Series5
Series6
Series7
0.4
Series8
Series9
0.2
0
6/1/01 7/1/01 8/1/01 9/1/01 10/1/01 11/1/01 12/1/01 1/1/02 2/1/02 3/1/02 4/1/02 5/1/02

19. 6/1/2002-5/31/2003
14
12
10
Series1
Series2
8 Series3
Series4
Series5
6 Series6
Series7
Series8
4
Series9
2
0
6/1/02 7/1/02 8/1/02 9/1/02 10/1/02 11/1/02 12/1/02 1/1/03 2/1/03 3/1/03 4/1/03 5/1/03

20. 6/1/2003-5/31/2004
30
25
20 Series1
Series2
Series3
Series4
15
Series5
Series6
Series7
10
Series8
Series9
5
0
6/1/03 7/1/03 8/1/03 9/1/03 10/1/03 11/1/03 12/1/03 1/1/04 2/1/04 3/1/04 4/1/04 5/1/04

21. 6/1/2004-5/31/2005
60
50
40 Series1
Series2
Series3
Series4
30
Series5
Series6
Series7
20
Series8
Series9
10
0
6/1/04 7/1/04 8/1/04 9/1/04 10/1/04 11/1/04 12/1/04 1/1/05 2/1/05 3/1/05 4/1/05 5/1/05

22. 6/1/2005-5/31/2006
30
25
20 Series1
Series2
Series3
Series4
15
Series5
Series6
Series7
10
Series8
Series9
5
0
6/1/05 7/1/05 8/1/05 9/1/05 10/1/05 11/1/05 12/1/05 1/1/06 2/1/06 3/1/06 4/1/06 5/1/06

23. 6/1/2006-5/31/2007
50
45
40
35
Series1
Series2
30
Series3
Series4
25
Series5
20 Series6
Series7
15 Series8
Series9
10
5
0
6/1/06 7/1/06 8/1/06 9/1/06 10/1/06 11/1/06 12/1/06 1/1/07 2/1/07 3/1/07 4/1/07 5/1/07

24. 6/1/2007-5/31/2008
400
350
300
Series1
250 Series2
Series3
Series4
200
Series5
Series6
150 Series7
Series8
100 Series9
50
0
6/1/07 7/1/07 8/1/07 9/1/07 10/1/07 11/1/07 12/1/07 1/1/08 2/1/08 3/1/08 4/1/08 5/1/08

25. 6/1/2008-5/31/2009
80
70
60
Series1
50 Series2
Series3
Series4
40
Series5
Series6
30 Series7
Series8
20 Series9
10
0
6/1/08 7/1/08 8/1/08 9/1/08 10/1/08 11/1/08 12/1/08 1/1/09 2/1/09 3/1/09 4/1/09 5/1/09

26. 6/1/2009-5/31/2010
200
180
160
140 Series1
Series2
120 Series3
Series4
100 Series5
Series6
80 Series7
Series8
60
Series9
Series10
40
20
0
6/1/09 7/1/09 8/1/09 9/1/09 10/1/09 11/1/09 12/1/09 1/1/10 2/1/10 3/1/10 4/1/10 5/1/10

27. Spike sizes over the years
400
350
300
250
200
Series2
150
100
50
0
2002 2003 2004 2005 2006 2007 2008 2009

28. Let’s take a closer look at the more
interesting actors

29. Actors #4-9 6/1/2008-5/31/2009
80
70
60
50
Series1
Series2
40 Series3
Series4
30 Series5
Series6
20
10
0
6/1/08 7/1/08 8/1/08 9/1/08 10/1/08 11/1/08 12/1/08 1/1/09 2/1/09 3/1/09 4/1/09 5/1/09

30. Actors #4-9 6/1/2008-5/31/2009 -log
2
1.8
1.6
1.4
1.2 Series1
Series2
1 Series3
Series4
0.8 Series5
Series6
0.6
0.4
0.2
0
6/1/08 7/1/08 8/1/08 9/1/08 10/1/08 11/1/08 12/1/08 1/1/09 2/1/09 3/1/09 4/1/09 5/1/09

31. One actor at a time ~10 years

32. Actor #1 DoD: 6/27/2001 -edits/day
14
12
10
8
Series1
6
4
2
0
6/28/01 6/28/02 6/28/03 6/28/04 6/28/05 6/28/06 6/28/07 6/28/08 6/28/09 6/28/10 6/28/11

33. Actor #1 –log(edits)/day
1.2
1
0.8
0.6
Series1
0.4
0.2
0
6/28/01 6/28/02 6/28/03 6/28/04 6/28/05 6/28/06 6/28/07 6/28/08 6/28/09 6/28/10 6/28/11

34. Actor #7 -edits/day
100
90
80
70
60
50
Series1
40
30
20
10
0
9/24/03 9/24/04 9/24/05 9/24/06 9/24/07 9/24/08 9/24/09 9/24/10 9/24/11

35. Actor #7 –log(edits)/day
2.5
2
1.5
Series1
1
0.5
0
9/24/03 9/24/04 9/24/05 9/24/06 9/24/07 9/24/08 9/24/09 9/24/10 9/24/11

36. Actor #8 -edits/day
400
350
300
250
200
Series1
150
100
50
0
12/10/03 12/10/04 12/10/05 12/10/06 12/10/07 12/10/08 12/10/09 12/10/10

37. Actor #8 –log(edits)/day
3
2.5
2
1.5
Series1
1
0.5
0
12/10/03 12/10/04 12/10/05 12/10/06 12/10/07 12/10/08 12/10/09 12/10/10

38. Actor #9 –edits/day
200
180
160
140
120
100
Series1
80
60
40
20
0
2/28/04 2/28/05 2/28/06 2/28/07 2/28/08 2/28/09 2/28/10 2/28/11

39. Actor #9 –log(edits)/day
2.5
2
1.5
Series1
1
0.5
0
2/28/04 2/28/05 2/28/06 2/28/07 2/28/08 2/28/09 2/28/10 2/28/11

40. If we tweak the data to take
importance into consideration…
 Average gross, adjusted for inflation*
 Only available for a small amount of actors chosen in the
sample set
 Taken from boxofficemojo.com
 Extremely reliable source

41. 50
100
150
200
250
300
350
400
0
1
9
17
25
33
41
49
57
65
73
81
89
97
105
113
121
129
137
145
153
161
169
177
185
193
201
209
217
225
233
241
249
Actor #8 vs. Actor #9
257
265
273
281
289
297
305
313
321
329
337
345
353
361
ledger
swayze

42. 50
100
150
200
250
300
350
400
0
1
9
17
25
33
41
49
57
65
73
81
89
97
105
113
121
129
137
145
153
161
169
177
185
193
201
209
217
225
233
241
249
257
265
273
281
289
297
305
313
321
329
337
345
353
361
Actor #8 vs. Actor #9 (adjusted)
ledger
swayze adjusted

43. 1.5
2.5
0.5
0
1
2
3
1
10
19
28
37
46
55
64
73
82
91
100
109
118
127
136
145
154
163
172
181
190
199
208
217
226
235
244
253
262
271
280
289
298
307
316
325
334
343
352
361
Actor #8 Vs. Actor #9 (adjusted)
ledger log
swayze adjusted log

44. The same data on Google trends

45. -10 days to +40 days (log)
3
2.5
2 coburn log
peck log
brando log
1.5 davis log
palance log
goulet log
1 ledger log
swayze log
0.5
0
1 2 3 4 5 6 7 8 9 1011121314151617181920212223242526272829303132333435363738394041424344454647484950

46. Other things I should consider
 Age at death
 Cause of death
 Were they still acting?

47. Future directions
 New sample of Wikipedia pages
 Need to compare more contemporary pages
 Need new metrics for comparison
 Better workflows

48. Thanks!
 Questions?
 http://www.slideshare.net/mlincol2/informetrics