Nicholas Christakis: How social networks predict epidemics

93,599 views ใƒป 2010-09-16

TED


ืื ื ืœื—ืฅ ืคืขืžื™ื™ื ืขืœ ื”ื›ืชื•ื‘ื™ื•ืช ื‘ืื ื’ืœื™ืช ืœืžื˜ื” ื›ื“ื™ ืœื”ืคืขื™ืœ ืืช ื”ืกืจื˜ื•ืŸ.

ืžืชืจื’ื: Hadas Shema ืžื‘ืงืจ: Ido Dekkers
00:15
For the last 10 years, I've been spending my time trying to figure out
0
15260
3000
ื‘ืžื”ืœืš ืขืฉืจ ื”ืฉื ื™ื ื”ืื—ืจื•ื ื•ืช, ื‘ื™ืœื™ืชื™ ืืช ื–ืžื ื™ ื‘ื ื™ืกื™ื•ืŸ ืœื”ื‘ื™ืŸ
00:18
how and why human beings
1
18260
2000
ื›ื™ืฆื“ ื•ืžื“ื•ืข ื‘ื ื™ ืื“ื
00:20
assemble themselves into social networks.
2
20260
3000
ืžืงื‘ืฆื™ื ืขืฆืžื ืœืจืฉืชื•ืช ื—ื‘ืจืชื™ื•ืช
00:23
And the kind of social network I'm talking about
3
23260
2000
ื•ืกื•ื’ื™ ื”ืจืฉืชื•ืช ื”ื—ื‘ืจืชื™ื•ืช ืฉืื ื™ ืžื“ื‘ืจ ืขืœื™ื”ืŸ
00:25
is not the recent online variety,
4
25260
2000
ืื™ื ืŸ ื”ืจืฉืชื•ืช ื”ืžืงื•ื•ื ื•ืช ืฉืฆืฆื• ืœืื—ืจื•ื ื”
00:27
but rather, the kind of social networks
5
27260
2000
ืืœื ืจืฉืชื•ืช ื—ื‘ืจืชื™ื•ืช ืžืกื•ื’
00:29
that human beings have been assembling for hundreds of thousands of years,
6
29260
3000
ืฉื‘ื ื™ ื”ืื“ื ื™ื•ืฆืจื™ื ื›ื‘ืจ ืžืื•ืช ืืœืคื™ ืฉื ื™ื
00:32
ever since we emerged from the African savannah.
7
32260
3000
ืžืื– ื”ื’ื—ื ื• ืžื”ืกื•ื•ืื ื” ื”ืืคืจื™ืงื ื™ืช
00:35
So, I form friendships and co-worker
8
35260
2000
ื›ืš, ืื ื™ ื™ื•ืฆืจ ื—ื‘ืจื•ื™ื•ืช ื•ืงืฉืจื™ ืขื‘ื•ื“ื”
00:37
and sibling and relative relationships with other people
9
37260
3000
ืงืฉืจื™ ืื—ื™ื ื•ืžืขืจื›ื•ืช ื™ื—ืกื™ื ืขื ื‘ื ื™ ืžืฉืคื—ื” ืื—ืจื™ื
00:40
who in turn have similar relationships with other people.
10
40260
2000
ื•ืœื”ื ื‘ืชื•ืจื ืžืขืจื›ื•ืช ื™ื—ืกื™ื ื“ื•ืžื•ืช ืขื ืื ืฉื™ื ืื—ืจื™ื.
00:42
And this spreads on out endlessly into a distance.
11
42260
3000
ื•ื–ื” ืžืชืคืฉื˜ ืขื“ ืื™ืŸ ืกื•ืฃ ืœืžืจื—ืงื™ื
00:45
And you get a network that looks like this.
12
45260
2000
ื•ืื ื• ืžืงื‘ืœื™ื ืจืฉืช ืฉื ืจืื™ืช ื›ืš.
00:47
Every dot is a person.
13
47260
2000
ื›ืœ ื ืงื•ื“ื” ืžื™ื™ืฆื’ืช ืื“ื
00:49
Every line between them is a relationship between two people --
14
49260
2000
ื›ืœ ืงื• ื‘ื™ื ื™ื”ืŸ ืžื™ื™ืฆื’ ืžืขืจื›ืช ื™ื—ืกื™ื ื‘ื™ืŸ ืฉื ื™ ื‘ื ื™ ืื“ื --
00:51
different kinds of relationships.
15
51260
2000
ืžืขืจื›ื•ืช ื™ื—ืกื™ื ืžืกื•ื’ื™ื ืฉื•ื ื™ื.
00:53
And you can get this kind of vast fabric of humanity,
16
53260
3000
ื•ืืชื ื™ื›ื•ืœื™ื ืœืงื‘ืœ ืžื™ืŸ ืžืืจื’ ืขืฆื•ื ืฉืœ ื”ืื ื•ืฉื•ืช,
00:56
in which we're all embedded.
17
56260
2000
ื‘ืชื•ื›ื• ื›ื•ืœื ื• ืžื•ื˜ืžืขื™ื.
00:58
And my colleague, James Fowler and I have been studying for quite sometime
18
58260
3000
ื•ืขืžื™ืชื™, ื’'ื™ื™ืžืก ืคืื•ืœืจ, ื•ืื ื™ ื—ื•ืงืจื™ื ื›ื‘ืจ ื–ืžืŸ ืจื‘ ืœืžื“ื™
01:01
what are the mathematical, social,
19
61260
2000
ืžื”ื ื”ื—ื•ืงื™ื ื”ืžืชืžื˜ื™ื™ื, ื”ื—ื‘ืจืชื™ื™ื
01:03
biological and psychological rules
20
63260
3000
ื”ื‘ื™ื•ืœื•ื’ื™ื™ื ื•ื”ืคืกื™ื›ื•ืœื•ื’ื™ื™ื
01:06
that govern how these networks are assembled
21
66260
2000
ื”ืžื•ืฉืœื™ื ื‘ื™ืฆื™ืจืช ื”ืจืฉืชื•ืช ื”ืœืœื•
01:08
and what are the similar rules
22
68260
2000
ื•ืžื”ื ื”ื—ื•ืงื™ื ื”ืžืงื‘ื™ืœื™ื
01:10
that govern how they operate, how they affect our lives.
23
70260
3000
ื”ืžื•ืฉืœื™ื ื‘ืื•ืคืŸ ืคืขื•ืœืชืŸ, ื‘ืื•ืคืŸ ื‘ื• ื”ืŸ ืžืฉืคื™ืขื•ืช ืขืœ ื—ื™ื™ื ื•.
01:13
But recently, we've been wondering
24
73260
2000
ื•ืœืื—ืจื•ื ื”, ื”ืชื—ืœื ื• ืœืชื”ื•ืช
01:15
whether it might be possible to take advantage of this insight,
25
75260
3000
ื”ืื ื™ื™ืชื›ืŸ ืœื ืฆืœ ืืช ื”ืชื•ื‘ื ื” ื”ื–ื•,
01:18
to actually find ways to improve the world,
26
78260
2000
ืขืœ ืžื ืช ืœืžืฆื•ื ื“ืจื›ื™ื ืœืฉืคืจ ืืช ื”ืขื•ืœื,
01:20
to do something better,
27
80260
2000
ืœืขืฉื™ื™ื” ื—ื™ื•ื‘ื™ืช ื™ื•ืชืจ,
01:22
to actually fix things, not just understand things.
28
82260
3000
ืœืชืงืŸ ื“ื‘ืจื™ื, ืœื ืจืง ืœื”ื‘ื™ืŸ ื“ื‘ืจื™ื.
01:25
So one of the first things we thought we would tackle
29
85260
3000
ืื– ืื—ื“ ื”ื“ื‘ืจื™ื ื”ืจืืฉื•ื ื™ื ืฉื—ืฉื‘ื ื• ืฉื ืชืžื•ื“ื“ ืื™ืชื
01:28
would be how we go about predicting epidemics.
30
88260
3000
ื”ื™ื” ื”ืื•ืคืŸ ื‘ื• ืื ื• ื—ื•ื–ื™ื ืžื’ื™ืคื•ืช.
01:31
And the current state of the art in predicting an epidemic --
31
91260
2000
ื•ื”ืฉื™ื˜ื” ื”ืขื“ื›ื ื™ืช ื‘ื™ื•ืชืจ ื‘ื ื™ื‘ื•ื™ ืžื’ื™ืคื” --
01:33
if you're the CDC or some other national body --
32
93260
3000
ืื ืืชื ื”ืžืจื›ื– ื”ืœืื•ืžื™ ืœื‘ืงืจืช ืžื—ืœื•ืช ืื• ื’ื•ืฃ ืœืื•ืžื™ ืื—ืจ ื›ืœืฉื”ื• --
01:36
is to sit in the middle where you are
33
96260
2000
ื”ื™ื ืœืฉื‘ืช ื‘ืžืจื›ื–, ืื™ืคื” ืฉืœื ืชื”ื™ื”,
01:38
and collect data
34
98260
2000
ื•ืœืืกื•ืฃ ื ืชื•ื ื™ื
01:40
from physicians and laboratories in the field
35
100260
2000
ืžืจื•ืคืื™ื ื•ืžืขื‘ื“ื•ืช ื”ื ืžืฆืื™ื ื‘ืฉื˜ื—
01:42
that report the prevalence or the incidence of certain conditions.
36
102260
3000
ื”ืžื“ื•ื•ื—ื™ื ืžื”ื™ ืฉื›ื™ื—ื•ืชื ืื• ื”ื™ืงืคื ืฉืœ ืชื ืื™ื ืžืกื•ื™ื™ืžื™ื.
01:45
So, so and so patients have been diagnosed with something,
37
105260
3000
ืื–, ื›ืš ื•ื›ืš ืžื˜ื•ืคืœื™ื ืื•ื‘ื—ื ื• ื›ื—ื•ืœื™ื ื‘ืžืฉื”ื• ืžืกื•ื™ื™ื [ื‘ืžืงื•ื ื–ื”]
01:48
or other patients have been diagnosed,
38
108260
2000
ืื• ืžื˜ื•ืคืœื™ื ืื—ืจื™ื ืื•ื‘ื—ื ื• ื›ื—ื•ืœื™ื [ื‘ืžืงื•ื ืื—ืจ]
01:50
and all these data are fed into a central repository, with some delay.
39
110260
3000
ื•ื›ืœ ื”ื ืชื•ื ื™ื ื”ืืœื• ืžื•ื–ื ื™ื ืœืžืื’ืจ ืžืจื›ื–ื™, ื‘ืื™ื—ื•ืจ ืžืกื•ื™ื™ื.
01:53
And if everything goes smoothly,
40
113260
2000
ื•ืื ื”ื›ืœ ืžืชื ื”ืœ ื›ืฉื•ืจื”
01:55
one to two weeks from now
41
115260
2000
ื‘ืขื•ื“ ืฉื‘ื•ืข ืขื“ ืฉื‘ื•ืขื™ื™ื ืžืขื›ืฉื™ื•,
01:57
you'll know where the epidemic was today.
42
117260
3000
ืชื“ืขื• ืื™ืคื” ื ืžืฆืืช ื”ืžื’ื™ืคื” ื”ื™ื•ื.
02:00
And actually, about a year or so ago,
43
120260
2000
ื•ืœืžืขืฉื”, ืœืคื ื™ ื›ืฉื ื”,
02:02
there was this promulgation
44
122260
2000
ื”ื™ื™ืชื” ืื•ืชื” ื”ื›ืจื–ื”
02:04
of the idea of Google Flu Trends, with respect to the flu,
45
124260
3000
ืขืœ ืจืขื™ื•ืŸ "ืžื’ืžื•ืช ื”ืฉืคืขืช ืฉืœ ื’ื•ื’ืœ", ื‘ื”ืชื™ื™ื—ืก ืœืฉืคืขืช,
02:07
where by looking at people's searching behavior today,
46
127260
3000
ืฉื, ื‘ืืžืฆืขื•ืช ื”ืชื‘ื•ื ื ื•ืช ื‘ื”ืชื ื”ื’ื•ืช ื”ื—ื™ืคื•ืฉ ืฉืœ ืื ืฉื™ื ื”ื™ื•ื
02:10
we could know where the flu --
47
130260
2000
ื ื•ื›ืœ ืœื“ืขืช ืื™ืคื” ื”ืฉืคืขืช...
02:12
what the status of the epidemic was today,
48
132260
2000
ืžื” ืžืฆื‘ื” ื”ืขื“ื›ื ื™ ืฉืœ ื”ืžื’ื™ืคื” ื”ื™ื•ื
02:14
what's the prevalence of the epidemic today.
49
134260
3000
ืžื” ืฉื›ื™ื—ื•ืช ื”ืžื’ื™ืคื” ื”ื™ื•ื
02:17
But what I'd like to show you today
50
137260
2000
ืื‘ืœ ืžื” ืฉื”ื™ื™ืชื™ ืจื•ืฆื” ืœื”ืจืื•ืช ืœื›ื ื”ื™ื•ื
02:19
is a means by which we might get
51
139260
2000
ื–ื”ื• ืืžืฆืขื™ ืฉื‘ืขื–ืจืชื• ื™ื™ืชื›ืŸ ืฉื ื•ื›ืœ ืœืงื‘ืœ
02:21
not just rapid warning about an epidemic,
52
141260
3000
ืœื ืจืง ืื–ื”ืจื” ื‘ื–ืžืŸ ืืžืช ื‘ืžืงืจื” ืฉืœ ืžื’ื™ืคื”
02:24
but also actually
53
144260
2000
ืืœื ืืคื™ืœื•
02:26
early detection of an epidemic.
54
146260
2000
ืื™ืชื•ืจ ืžื•ืงื“ื ืฉืœ ืžื’ืคื”.
02:28
And, in fact, this idea can be used
55
148260
2000
ื•ืœืžืขืฉื”, ื”ืจืขื™ื•ืŸ ื”ื–ื” ื™ื›ื•ืœ ืœืฉืžืฉ
02:30
not just to predict epidemics of germs,
56
150260
3000
ืœื ืจืง ืœืฉื ื—ื™ื–ื•ื™ ืžื’ืคื•ืช ื—ื™ื™ื“ืงื™ื•ืช
02:33
but also to predict epidemics of all sorts of kinds.
57
153260
3000
ืืœื ื’ื ืžื’ื™ืคื•ืช ืžืžื’ื•ื•ืŸ ืกื•ื’ื™ื.
02:37
For example, anything that spreads by a form of social contagion
58
157260
3000
ืœื“ื•ื’ืžื”, ื›ืœ ืžื” ืฉืžืชืคืฉื˜ ื‘ืฆื•ืจื” ืฉืœ ื”ื“ื‘ืงื” ื—ื‘ืจืชื™ืช
02:40
could be understood in this way,
59
160260
2000
ื™ื•ื›ืœ ืœื”ื™ื•ืช ืžื•ื‘ืŸ ื‘ืฆื•ืจื” ื”ื–ื•,
02:42
from abstract ideas on the left
60
162260
2000
ื”ื—ืœ ืžืจืขื™ื•ื ื•ืช ืžื•ืคืฉื˜ื™ื ืžืฉืžืืœ
02:44
like patriotism, or altruism, or religion
61
164260
3000
ื›ืžื• ืคื˜ืจื™ื•ื˜ื™ื•ืช, ืืœื˜ืจื•ืื™ื–ื ืื• ื“ืช
02:47
to practices
62
167260
2000
ื•ื›ืœื” ื‘ืžื ื”ื’ื™ื
02:49
like dieting behavior, or book purchasing,
63
169260
2000
ื›ืžื• ื”ืจื’ืœื™ ื“ื™ืื˜ื”, ืื• ืงื ื™ื™ืช ืกืคืจื™ื,
02:51
or drinking, or bicycle-helmet [and] other safety practices,
64
171260
3000
ืื• ืฉืชื™ื™ื”, ืื• ืงืกื“ื•ืช ืื•ืคื ื™ื™ื...[ื•] ื ื•ื”ื’ื™ ื‘ื˜ื™ื—ื•ืช ืื—ืจื™ื,
02:54
or products that people might buy,
65
174260
2000
ืื• ืžื•ืฆืจื™ื ืฉืื ืฉื™ื ืขืฉื•ื™ื™ื ืœืงื ื•ืช,
02:56
purchases of electronic goods,
66
176260
2000
ืจื›ื™ืฉื•ืช ืฉืœ ืžื•ืฆืจื™ ืืœืงื˜ืจื•ื ื™ืงื”
02:58
anything in which there's kind of an interpersonal spread.
67
178260
3000
ื›ืœ ื“ื‘ืจ ื‘ื• ื™ืฉ ืฆื•ืจื” ื›ืœืฉื”ื™ ืฉืœ ื”ืชืคืฉื˜ื•ืช ื‘ื™ืŸ-ืื™ืฉื™ืช.
03:01
A kind of a diffusion of innovation
68
181260
2000
ืžืขื™ืŸ ื“ื™ืคื•ื–ื™ื” ืฉืœ ื—ื“ืฉื ื•ืช
03:03
could be understood and predicted
69
183260
2000
ื”ื ื™ืชื ืช ืœื”ื‘ื ื” ื•ืœื ื™ื‘ื•ื™
03:05
by the mechanism I'm going to show you now.
70
185260
3000
ื‘ืขื–ืจืช ื”ืžื ื’ื ื•ืŸ ืฉืื ื™ ืขื•ืžื“ ืœื”ืจืื•ืช ืœื›ื ื›ืขืช.
03:08
So, as all of you probably know,
71
188260
2000
ืื–, ื›ืžื• ืฉื›ื•ืœื›ื ื‘ื•ื•ื“ืื™ ื™ื•ื“ืขื™ื,
03:10
the classic way of thinking about this
72
190260
2000
ื”ื“ืจืš ื”ืงืœืืกื™ืช ืœื—ืฉื•ื‘ ืขืœ ื–ื”
03:12
is the diffusion-of-innovation,
73
192260
2000
ื”ื™ื ื“ืจืš ื”ื”ืคืฆื” ืฉืœ ื—ื™ื“ื•ืฉ,
03:14
or the adoption curve.
74
194260
2000
ืื• ืขืงื•ืžืช ื”ื”ื˜ืžืขื”.
03:16
So here on the Y-axis, we have the percent of the people affected,
75
196260
2000
ืื– ื›ืืŸ, ืขืœ ืฆื™ืจ ื”-Y, ื™ืฉ ืœื ื• ืืช ืื—ื•ื– ื”ืื ืฉื™ื ื”ืžื•ืฉืคืขื™ื
03:18
and on the X-axis, we have time.
76
198260
2000
ื•ืขืœ ืฆื™ืจ ื”-X, ื™ืฉ ืœื ื• ื–ืžืŸ.
03:20
And at the very beginning, not too many people are affected,
77
200260
3000
ืžืžืฉ ื‘ื”ืชื—ืœื”, ืœื ื™ื•ืชืจ ืžื“ื™ ืื ืฉื™ื ืžื•ืฉืคืขื™ื,
03:23
and you get this classic sigmoidal,
78
203260
2000
ื•ืื ื• ืžืงื‘ืœื™ื ืขืงื•ืžื” ืกื™ื’ืžื•ืื™ื“ื™ืช,
03:25
or S-shaped, curve.
79
205260
2000
ืื• ืขืงื•ืžื” ื“ืžื•ื™ื™ืช-S.
03:27
And the reason for this shape is that at the very beginning,
80
207260
2000
ื•ื”ืกื™ื‘ื” ืœืฆื•ืจื” ื”ื–ื• ื”ื™ื ืฉืžืžืฉ ื‘ื”ืชื—ืœื”,
03:29
let's say one or two people
81
209260
2000
ื‘ื•ืื• ื ืืžืจ ืื“ื ืื—ื“ ืื• ืฉื ื™ื™ื
03:31
are infected, or affected by the thing
82
211260
2000
ืžื•ืฉืคืขื™ื, ืื• ืžื•ื“ื‘ืงื™ื, ืขืœ ื™ื“ื™ ื”ื“ื‘ืจ ื”ืžืกื•ื™ื™ื,
03:33
and then they affect, or infect, two people,
83
213260
2000
ื•ื”ื ืžืฉืคื™ืขื™ื ืขืœ, ืื• ืžื“ื‘ื™ืงื™ื, ืฉื ื™ ืื ืฉื™ื,
03:35
who in turn affect four, eight, 16 and so forth,
84
215260
3000
ืฉื‘ืชื•ืจื ืžืฉืคื™ืขื™ื ืขืœ ืืจื‘ืขื”, ืฉืžื•ื ื”, ืฉื™ืฉื” ืขืฉืจ ื•ื›ืŸ ื”ืœืื”,
03:38
and you get the epidemic growth phase of the curve.
85
218260
3000
ื•ืื– ืžืงื‘ืœื™ื ืืช ืฉืœื‘ ื”ืชืคืฉื˜ื•ืช ื”ืžื’ื™ืคื” ืฉืœ ื”ืขืงื•ืžื”.
03:41
And eventually, you saturate the population.
86
221260
2000
ื•ืœื‘ืกื•ืฃ, ืืชื ืžืจื•ื•ื™ื ืืช ื”ืื•ื›ืœื•ืกื™ื”.
03:43
There are fewer and fewer people
87
223260
2000
ื™ืฉ ืคื—ื•ืช ื•ืคื—ื•ืช ืื ืฉื™ื
03:45
who are still available that you might infect,
88
225260
2000
ืฉืขื“ื™ื™ืŸ ื–ืžื™ื ื™ื ืœื”ื“ื‘ืงื”,
03:47
and then you get the plateau of the curve,
89
227260
2000
ื•ืื– ืžืงื‘ืœื™ื ืืช ื”ื™ืฉื•ืจ ืฉืœ ื”ืขืงื•ืžื”,
03:49
and you get this classic sigmoidal curve.
90
229260
3000
ื•ืืชื ืžืงื‘ืœื™ื ืืช ื”ืขืงื•ืžื” ื”ืกื™ื’ืžื•ืื™ื“ื™ืช ื”ืงืœืืกื™ืช.
03:52
And this holds for germs, ideas,
91
232260
2000
ื•ื–ื” ื ื›ื•ืŸ ื‘ืขื‘ื•ืจ ื—ื™ื™ื“ืงื™ื, ืจืขื™ื•ื ื•ืช
03:54
product adoption, behaviors,
92
234260
2000
ืื™ืžื•ืฅ ืฉืœ ืžื•ืฆืจื™ื, ื”ืชื ื”ื’ื•ื™ื•ืช
03:56
and the like.
93
236260
2000
ื•ื“ื•ืžื™ื”ื.
03:58
But things don't just diffuse in human populations at random.
94
238260
3000
ืืœื ืฉื“ื‘ืจื™ื ืื™ื ื ืžืชืคืฉื˜ื™ื ื‘ืื•ื›ืœื•ืกื™ื•ืช ืื ื•ืฉื™ื•ืช ื‘ืื•ืคืŸ ืืงืจืื™.
04:01
They actually diffuse through networks.
95
241260
2000
ื”ื ืœืžืขืฉื” ืžืชืคืฉื˜ื™ื ื“ืจืš ืจืฉืชื•ืช.
04:03
Because, as I said, we live our lives in networks,
96
243260
3000
ืžืื—ืจ ื•ื›ืžื• ืฉืืžืจืชื™, ืื ื• ื—ื™ื™ื ืืช ื—ื™ื™ื ื• ื‘ืจืฉืชื•ืช,
04:06
and these networks have a particular kind of a structure.
97
246260
3000
ื•ืœืจืฉืชื•ืช ืืœื• ื™ืฉ ืžื‘ื ื” ืžืกื•ื’ ืžืกื•ื™ื.
04:09
Now if you look at a network like this --
98
249260
2000
ืื ื ืกืชื›ืœ ืขืœ ืจืฉืช ื›ืžื• ื–ื•...
04:11
this is 105 people.
99
251260
2000
ื™ืฉ ื›ืืŸ 105 ืื ืฉื™ื.
04:13
And the lines represent -- the dots are the people,
100
253260
2000
ื•ื”ืงื•ื•ื™ื ืžื™ื™ืฆื’ื™ื.. ื”ื ืงื•ื“ื•ืช ื”ื ื”ืื ืฉื™ื,
04:15
and the lines represent friendship relationships.
101
255260
2000
ื•ื”ืงื•ื•ื™ื ืžื™ื™ืฆื’ื™ื ืงืฉืจื™ ื—ื‘ืจื•ืช.
04:17
You might see that people occupy
102
257260
2000
ืื•ืœื™ ืชื•ื›ืœื• ืœืจืื•ืช ืฉืื ืฉื™ื ืžืื›ืœืกื™ื
04:19
different locations within the network.
103
259260
2000
ืžื™ืงื•ืžื™ื ืฉื•ื ื™ื ื‘ืชื•ืš ื”ืจืฉืช.
04:21
And there are different kinds of relationships between the people.
104
261260
2000
ื•ื™ืฉื ื ืกื•ื’ื™ื ืฉื•ื ื™ื ืฉืœ ืžืขืจื›ื•ืช ื™ื—ืกื™ื ื‘ื™ืŸ ื”ืื ืฉื™ื.
04:23
You could have friendship relationships, sibling relationships,
105
263260
3000
ื™ื›ื•ืœื™ื ืœื”ื™ื•ืช ืœื›ื ืžืขืจื›ื•ืช ื™ื—ืกื™ื ืขื ื—ื‘ืจื™ื, ืžืขืจื›ื•ืช ื™ื—ืกื™ื ืขื ืื—ื™ื,
04:26
spousal relationships, co-worker relationships,
106
266260
3000
ืžืขืจื›ื•ืช ื™ื—ืกื™ื ืขื ื‘ื ื™ ื–ื•ื’, ืžืขืจื›ื•ืช ื™ื—ืกื™ื ืขื ื—ื‘ืจื™ื ืœืขื‘ื•ื“ื”.
04:29
neighbor relationships and the like.
107
269260
3000
ืžืขืจื›ื•ืช ื™ื—ืกื™ื ืขื ืฉื›ื ื™ื ื•ื›ื“ื•ืžื”.
04:32
And different sorts of things
108
272260
2000
ื•ื“ื‘ืจื™ื ืžืกื•ื’ื™ื ืฉื•ื ื™ื
04:34
spread across different sorts of ties.
109
274260
2000
ืžืชืคืฉื˜ื™ื ื“ืจืš ืงืฉืจื™ื ืžืกื•ื’ื™ื ืฉื•ื ื™ื.
04:36
For instance, sexually transmitted diseases
110
276260
2000
ืœืžืฉืœ, ืžื—ืœื•ืช ืžื™ืŸ
04:38
will spread across sexual ties.
111
278260
2000
ื™ืชืคื–ืจื• ืœืื•ืจืš ืงืฉืจื™ื ืžื™ื ื™ื™ื
04:40
Or, for instance, people's smoking behavior
112
280260
2000
ืื• ืœื“ื•ื’ืžื”, ื”ืจื’ืœื™ ืขื™ืฉื•ืŸ ืฉืœ ืื ืฉื™ื
04:42
might be influenced by their friends.
113
282260
2000
ืขืœื•ืœื™ื ืœื”ื™ื•ืช ืžื•ืฉืคืขื™ื ืžื—ื‘ืจื™ื”ื.
04:44
Or their altruistic or their charitable giving behavior
114
284260
2000
ืื• ื”ืชื ื”ื’ื•ืชื ื”ืืœื˜ืจื•ืื™ืกื˜ื™ืช ืื• ื”ื ืชื™ื ื” ืฉืœื”ื ืœืฆื“ืงื”
04:46
might be influenced by their coworkers,
115
286260
2000
ืขืฉื•ื™ื™ื ืœื”ื™ื•ืช ืžื•ืฉืคืขื™ื ืขืœ ื™ื“ื™ ื—ื‘ืจื™ื”ื ืœืขื‘ื•ื“ื”,
04:48
or by their neighbors.
116
288260
2000
ืื• ืขืœ ื™ื“ื™ ืฉื›ื ื™ื”ื.
04:50
But not all positions in the network are the same.
117
290260
3000
ืื‘ืœ ืœื ื›ืœ ื”ืขืžื“ื•ืช ื‘ืจืฉืช ื“ื•ืžื•ืช ื–ื• ืœื–ื•.
04:53
So if you look at this, you might immediately grasp
118
293260
2000
ืื– ืื ืชืกืชื›ืœื• ื‘ืชืžื•ื ื”, ื™ื™ืชื›ืŸ ืฉืชืชืคืกื• ืžื™ื“
04:55
that different people have different numbers of connections.
119
295260
3000
ืฉืœืื ืฉื™ื ืฉื•ื ื™ื ื™ืฉ ืžืกืคืจ ืฉื•ื ื” ืฉืœ ืงืฉืจื™ื.
04:58
Some people have one connection, some have two,
120
298260
2000
ืœื—ืœืง ืžื”ืื ืฉื™ื ื™ืฉ ืงืฉืจ ืื—ื“, ืœืื—ืจื™ื ื™ืฉ ืฉื ื™ื™ื.
05:00
some have six, some have 10 connections.
121
300260
3000
ืœื—ืœืง ื™ืฉ ืฉื™ืฉื”, ืœืื—ืจื™ื ืขืฉืจื” ืงืฉืจื™ื.
05:03
And this is called the "degree" of a node,
122
303260
2000
ื•ื–ื” ื ืงืจื” "ืžืขืœื”" ืฉืœ ืฆื•ืžืช,
05:05
or the number of connections that a node has.
123
305260
2000
ืื• ืžืกืคืจ ื”ืงืฉืจื™ื ืฉื™ืฉ ืœืฆื•ืžืช.
05:07
But in addition, there's something else.
124
307260
2000
ืืš ื‘ื ื•ืกืฃ, ื™ืฉ ืžืฉื”ื• ืื—ืจ.
05:09
So, if you look at nodes A and B,
125
309260
2000
ื›ืš, ืื ืชืกืชื›ืœื• ื‘ืฆืžืชื™ื A ื•-B,
05:11
they both have six connections.
126
311260
2000
ืœื›ืœ ืื—ื“ ื™ืฉ ืฉื™ืฉื” ืงืฉืจื™ื.
05:13
But if you can see this image [of the network] from a bird's eye view,
127
313260
3000
ืื‘ืœ ืื ืืชื ืžืกื•ื’ืœื™ื ืœืจืื•ืช ืืช ื”ืชืžื•ื ื” ื”ื–ื• [ืฉืœ ื”ืจืฉืช] ืžืžืขื•ืฃ ื”ืฆื™ืคื•ืจ,
05:16
you can appreciate that there's something very different
128
316260
2000
ืืชื ื™ื›ื•ืœื™ื ืœื”ืขืจื™ืš ืฉื™ืฉ ืžืฉื”ื• ืฉื•ื ื” ืžืื•ื“
05:18
about nodes A and B.
129
318260
2000
ื‘ื™ืŸ ืฆืžืชื™ื A ื•-B.
05:20
So, let me ask you this -- I can cultivate this intuition by asking a question --
130
320260
3000
ืื–, ืชื ื• ืœื™ ืœืฉืื•ืœ ืืชื›ื ืืช ื–ื” -- ืื ืกื” ืœืคืชื— ืืช ื”ืื™ื ื˜ื•ืื™ืฆื™ื” ื”ื–ื• ืขืœ ื™ื“ื™ ืฉืืœื” -
05:23
who would you rather be
131
323260
2000
ืžื™ ื”ื™ื™ืชื ืžืขื“ื™ืคื™ื ืœื”ื™ื•ืช
05:25
if a deadly germ was spreading through the network, A or B?
132
325260
3000
ืื ื—ื™ื™ื“ืง ืงื˜ืœื ื™ ื”ื™ื” ืžืชืคืฉื˜ ื“ืจืš ื”ืจืฉืช, A ืื• B ?
05:28
(Audience: B.) Nicholas Christakis: B, it's obvious.
133
328260
2000
(ืงื”ืœ : B.) ื ื™ืงื•ืœืก ื›ืจื™ืกื˜ืืงื™ืก: B, ื–ื” ื‘ืจื•ืจ.
05:30
B is located on the edge of the network.
134
330260
2000
B ืžืžื•ืงื ื‘ืงืฆื” ื”ืจืฉืช.
05:32
Now, who would you rather be
135
332260
2000
ืขื›ืฉื™ื•, ืžื™ ื”ื™ื™ืชื ืžืขื“ื™ืคื™ื ืœื”ื™ื•ืช
05:34
if a juicy piece of gossip were spreading through the network?
136
334260
3000
ืื ืคื™ืกืช ืจื›ื™ืœื•ืช ืขืกื™ืกื™ืช ื”ื™ื™ืชื” ืžืชืคืฉื˜ืช ื“ืจืš ื”ืจืฉืช?
05:37
A. And you have an immediate appreciation
137
337260
3000
A. ื•ื™ืฉ ืœื›ื ื™ื›ื•ืœืช ืœื”ืขืจื›ื” ืžื™ื™ื“ื™ืช
05:40
that A is going to be more likely
138
340260
2000
ืฉืกื‘ื™ืจ ื™ื•ืชืจ ืœื”ื ื™ื— ืฉ-A
05:42
to get the thing that's spreading and to get it sooner
139
342260
3000
ื™ื—ืฉืฃ ืœืžื” ืฉืžืชืคืฉื˜ ื•ื™ื—ืฉืฃ ืืœื™ื• ืžื•ืงื“ื ื™ื•ืชืจ
05:45
by virtue of their structural location within the network.
140
345260
3000
ื‘ื–ื›ื•ืช ืžื™ืงื•ืžื• ื”ืžื‘ื ื™ ื‘ืชื•ืš ื”ืจืฉืช.
05:48
A, in fact, is more central,
141
348260
2000
A, ืœืžืขืฉื”, ื”ื•ื ืžืจื›ื–ื™ ื™ื•ืชืจ,
05:50
and this can be formalized mathematically.
142
350260
3000
ื•ืชื›ื•ื ื” ื–ื• ืืคืฉืจ ืœืืฉืจ ืจืฉืžื™ืช ื‘ืื•ืคืŸ ืžืชืžื˜ื™.
05:53
So, if we want to track something
143
353260
2000
ืื–, ืื ืื ื• ืจื•ืฆื™ื ืœืขืงื•ื‘ ืื—ืจ ืžืฉื”ื•
05:55
that was spreading through a network,
144
355260
3000
ืฉื”ืชืคืฉื˜ ื“ืจืš ืจืฉืช ื›ืœืฉื”ื™,
05:58
what we ideally would like to do is to set up sensors
145
358260
2000
ืžื” ืฉื”ื™ื™ื ื• ืจื•ืฆื™ื ืœืขืฉื•ืช, ื‘ืื•ืคืŸ ืื™ื“ื™ืืœื™, ื”ื•ื ืœื”ืฆื™ื‘ ื—ื™ื™ืฉื ื™ื
06:00
on the central individuals within the network,
146
360260
2000
ื‘ืฆืžืชื™ื ื”ืžืจื›ื–ื™ื™ื ื‘ืจืฉืช,
06:02
including node A,
147
362260
2000
ื›ื•ืœืœ ืฆื•ืžืช A,
06:04
monitor those people that are right there in the middle of the network,
148
364260
3000
ืœื ื˜ืจ ืื ืฉื™ื ืืœื•, ืฉื”ื ืžืžืฉ ื‘ืžืจื›ื– ื”ืจืฉืช,
06:07
and somehow get an early detection
149
367260
2000
ื•ื‘ืฆื•ืจื” ื›ืœืฉื”ื™ ืœืงื‘ืœ ืื™ืชื•ืจ ืžื•ืงื“ื
06:09
of whatever it is that is spreading through the network.
150
369260
3000
ืฉืœ ืžื” ืฉืœื ื™ื”ื™ื” ืฉืžืชืคืฉื˜ ื“ืจืš ื”ืจืฉืช..
06:12
So if you saw them contract a germ or a piece of information,
151
372260
3000
ื›ืœื•ืžืจ, ืื ื”ื‘ื—ื ืชื ื‘ื›ืš ืฉื”ื ื ื“ื‘ืงื™ื ื‘ื—ื™ื™ื“ืง ืื• ืžืชื•ื•ื“ืขื™ื ืœืžื™ื“ืข ื›ืœืฉื”ื•,
06:15
you would know that, soon enough,
152
375260
2000
ืชื•ื›ืœื• ืœื“ืขืช ืฉื‘ืงืจื•ื‘ ืžืื•ื“,
06:17
everybody was about to contract this germ
153
377260
2000
ื›ื•ืœื ืขื•ืžื“ื™ื ืœื”ื“ื‘ืง ื‘ื—ื™ื™ื“ืง ื”ื–ื”
06:19
or this piece of information.
154
379260
2000
ืื• ืœื’ืœื•ืช ืืช ื”ืžื™ื“ืข ื”ื–ื”.
06:21
And this would be much better
155
381260
2000
ื•ื–ื” ื™ื”ื™ื” ื”ืจื‘ื” ื™ื•ืชืจ ื˜ื•ื‘
06:23
than monitoring six randomly chosen people,
156
383260
2000
ืžืืฉืจ ืœื ื˜ืจ ืฉื™ืฉื” ืื ืฉื™ื ืฉื ื‘ื—ืจื• ื‘ืฆื•ืจื” ืืงืจืื™ืช,
06:25
without reference to the structure of the population.
157
385260
3000
ื‘ืœื™ ื”ืชื™ื™ื—ืกื•ืช ืœืžื‘ื ื” ื”ืื•ื›ืœื•ืกื™ื”.
06:28
And in fact, if you could do that,
158
388260
2000
ื•ืœืžืขืฉื”, ืื ื ื•ื›ืœ ืœืขืฉื•ืช ื–ืืช,
06:30
what you would see is something like this.
159
390260
2000
ืžื” ืฉืชืจืื• ื”ื•ื ืžืฉื”ื• ื›ืžื• ื–ื”.
06:32
On the left-hand panel, again, we have the S-shaped curve of adoption.
160
392260
3000
ื‘ืคืื ืœ ื”ืฉืžืืœื™ ื™ืฉ ืœื ื•, ืฉื•ื‘, ืืช ืขืงื•ืžืช ื”-S ืฉืœ ืื™ืžื•ืฅ ื”ื—ื™ื“ื•ืฉ.
06:35
In the dotted red line, we show
161
395260
2000
ื‘ืงื• ื”ืžื ื•ืงื“ ื‘ืื“ื•ื, ืื ื• ืžืจืื™ื
06:37
what the adoption would be in the random people,
162
397260
2000
ืžื” ื™ื”ื™ื” ื”ืื™ืžื•ืฅ ืฉืœ ื”ืื ืฉื™ื ืฉื ื‘ื—ืจื• ืืงืจืื™ืช
06:39
and in the left-hand line, shifted to the left,
163
399260
3000
ื•ื‘ืงื• ื”ืฉืžืืœื™ (ื”ืฆื”ื•ื‘), ื‘ื”ืกื˜ื” ืžืกื•ื™ื™ืžืช ืœืฉืžืืœ,
06:42
we show what the adoption would be
164
402260
2000
ืื ื• ืจื•ืื™ื ืืช ืขืงื•ืžืช ื”ืื™ืžื•ืฅ ืฉืœ
06:44
in the central individuals within the network.
165
404260
2000
ื”ืคืจื˜ื™ื ื”ืžืจื›ื–ื™ื™ื ื‘ืจืฉืช.
06:46
On the Y-axis is the cumulative instances of contagion,
166
406260
2000
ืฆื™ืจ ื”-Y ืžื™ื™ืฆื’ ืืช ื”ืฆื˜ื‘ืจื•ืช ืžืงืจื™ ื”ื”ื™ื“ื‘ืงื•ืช,
06:48
and on the X-axis is the time.
167
408260
2000
ื•ืฆื™ืจ ื”-X ืžื™ื™ืฆื’ ืืช ื”ื–ืžืŸ.
06:50
And on the right-hand side, we show the same data,
168
410260
2000
ื•ื‘ืฆื“ ื™ืžื™ืŸ, ืื ื• ืžืจืื™ื ืื•ืชื ื”ื ืชื•ื ื™ื
06:52
but here with daily incidence.
169
412260
2000
ืื‘ืœ ื‘ืžื“ื™ื“ื” ื™ื•ืžื™ืช.
06:54
And what we show here is -- like, here --
170
414260
2000
ื•ืžื” ืฉืื ื• ืžืจืื™ื ื›ืืŸ ื”ื•ื -- ื›ืื™ืœื•, ื›ืืŸ --
06:56
very few people are affected, more and more and more and up to here,
171
416260
2000
ืžืขื˜ ืžืื•ื“ ืื ืฉื™ื ืžื•ืฉืคืขื™ื, ืขื•ื“ ื•ืขื•ื“ ื•ืขื•ื“ ื•ืขื“ ืœืžืขืœื”,
06:58
and here's the peak of the epidemic.
172
418260
2000
ื•ื”ื ื” ืฉื™ื ื”ืžื’ืคื”.
07:00
But shifted to the left is what's occurring in the central individuals.
173
420260
2000
ืžื•ืกื˜ ืœืฉืžืืœ ื”ื•ื ืžื” ืฉืžืชืจื—ืฉ ืืฆืœ ื”ืคืจื˜ื™ื ื”ืžืจื›ื–ื™ื™ื.
07:02
And this difference in time between the two
174
422260
3000
ื•ื”ื”ื‘ื“ืœ ื”ื–ื” ื‘ื–ืžืŸ ื‘ื™ืŸ ื”ืฉื ื™ื™ื
07:05
is the early detection, the early warning we can get,
175
425260
3000
ื”ื•ื ื”ืื™ืชื•ืจ ื”ืžื•ืงื“ื, ื”ืื–ื”ืจื” ื”ืžื•ืงื“ืžืช ืฉืื ื• ื™ื›ื•ืœื™ื ืœืงื‘ืœ,
07:08
about an impending epidemic
176
428260
2000
ืขืœ ื”ืžื’ืคื” ืฉื‘ื“ืจืš
07:10
in the human population.
177
430260
2000
ื‘ืื•ื›ืœื•ืกื™ื™ื” ื”ืื ื•ืฉื™ืช.
07:12
The problem, however,
178
432260
2000
ืื•ืœื ื”ื‘ืขื™ื” ื”ื™ื,
07:14
is that mapping human social networks
179
434260
2000
ืฉืžื™ืคื•ื™ ืจืฉืชื•ืช ื—ื‘ืจืชื™ื•ืช ืฉืœ ื‘ื ื™ ืื“ื
07:16
is not always possible.
180
436260
2000
ืื™ื ื• ืชืžื™ื“ ืืคืฉืจื™.
07:18
It can be expensive, not feasible,
181
438260
2000
ื–ื” ื™ื›ื•ืœ ืœื”ื™ื•ืช ื™ืงืจ, [ืงืฉื” ืžืื•ื“],
07:20
unethical,
182
440260
2000
ืœื ืืชื™,
07:22
or, frankly, just not possible to do such a thing.
183
442260
3000
ืื•, ืœืžืขืŸ ื”ืืžืช, ืคืฉื•ื˜ ืœื ืืคืฉืจื™.
07:25
So, how can we figure out
184
445260
2000
ืื–, ื›ื™ืฆื“ ื ื‘ื™ืŸ
07:27
who the central people are in a network
185
447260
2000
ืžื™ ื”ื ื”ืื ืฉื™ื ื”ืžืจื›ื–ื™ื™ื ื‘ืจืฉืช
07:29
without actually mapping the network?
186
449260
3000
ืžื‘ืœื™ ืœืžืขืฉื” ืœืžืคื•ืช ืืช ื”ืจืฉืช ?
07:32
What we came up with
187
452260
2000
ืžื” ืฉื”ืขืœื™ื ื•
07:34
was an idea to exploit an old fact,
188
454260
2000
ื”ื•ื ืจืขื™ื•ืŸ ื”ืžื ืฆืœ ืขื•ื‘ื“ื” ื™ืฉื ื”,
07:36
or a known fact, about social networks,
189
456260
2000
ืื• ืขื•ื‘ื“ื” ื™ื“ื•ืขื”, ืขืœ ืจืฉืชื•ืช ื—ื‘ืจืชื™ื•ืช,
07:38
which goes like this:
190
458260
2000
ืฉื”ื•ืœืš ื›ืš:
07:40
Do you know that your friends
191
460260
2000
ื”ืื ืืชื” ื™ื•ื“ืข ืฉืœื—ื‘ืจื™ื ืฉืœืš
07:42
have more friends than you do?
192
462260
3000
ื™ืฉ ื™ื•ืชืจ ื—ื‘ืจื™ื ืžืืฉืจ ืœืš ?
07:45
Your friends have more friends than you do,
193
465260
3000
ืœื—ื‘ืจื™ื ืฉืœืš ื™ืฉ ื™ื•ืชืจ ื—ื‘ืจื™ื ืžืืฉืจ ืœืš.
07:48
and this is known as the friendship paradox.
194
468260
2000
ื•ื–ื” ื™ื“ื•ืข ื‘ืชื•ืจ ืคืจื“ื•ืงืก ื”ื—ื‘ืจื•ืช.
07:50
Imagine a very popular person in the social network --
195
470260
2000
ื“ืžื™ื™ื ื• ืœืขืฆืžื›ื ืื“ื ืžืื•ื“ ืคื•ืคื•ืœืจื™ ื‘ืจืฉืช ื—ื‘ืจืชื™ืช --
07:52
like a party host who has hundreds of friends --
196
472260
3000
ื›ืžื• ืžืืจื— ืžืกื™ื‘ื” ืฉื™ืฉ ืœื• ืžืื•ืช ื—ื‘ืจื™ื --
07:55
and a misanthrope who has just one friend,
197
475260
2000
ื•ืžื™ื–ื ื˜ืจื•ืค (ืฉื•ื ื ืื“ื) ืฉื™ืฉ ืœื• ืจืง ื—ื‘ืจ ืื—ื“,
07:57
and you pick someone at random from the population;
198
477260
3000
ื•ืชื‘ื—ืจื• ืžื™ืฉื”ื• ื‘ืื•ืคืŸ ืืงืจืื™ ืžื”ืื•ื›ืœื•ืกื™ื”;
08:00
they were much more likely to know the party host.
199
480260
2000
ืœื”ื ื™ืฉ ื”ืจื‘ื” ื™ื•ืชืจ ืกื™ื›ื•ื™ ืœื”ื›ื™ืจ ืืช ืžืืจื— ื”ืžืกื™ื‘ื”.
08:02
And if they nominate the party host as their friend,
200
482260
2000
ื•ืื ื”ื ื‘ื•ื—ืจื™ื ื‘ืžืืจื— ื”ืžืกื™ื‘ื” ื‘ืชื•ืจ ื”ื—ื‘ืจ ืฉืœื”ื,
08:04
that party host has a hundred friends,
201
484260
2000
ืœืžืืจื— ื™ืฉ ืžืื” ื—ื‘ืจื™ื,
08:06
therefore, has more friends than they do.
202
486260
3000
ื•ืœื›ืŸ ื™ืฉ ืœื• ื™ื•ืชืจ ื—ื‘ืจื™ื ืžืืฉืจ ืœื”ื.
08:09
And this, in essence, is what's known as the friendship paradox.
203
489260
3000
ื•ื–ื”, ื‘ืžื”ื•ืชื•, ื™ื“ื•ืข ื›ืคืจื“ื•ืงืก ื”ื—ื‘ืจื•ืช.
08:12
The friends of randomly chosen people
204
492260
3000
ื”ื—ื‘ืจื™ื ืฉืœ ืื ืฉื™ื ืฉื ื‘ื—ืจื• ื‘ืืงืจืื™
08:15
have higher degree, and are more central
205
495260
2000
ื‘ืขืœื™ ื“ืจื’ื” ื™ื•ืชืจ ื’ื‘ื•ื”ื”, ื•ื”ื ื™ื•ืชืจ ืžืจื›ื–ื™ื™ื,
08:17
than the random people themselves.
206
497260
2000
ืžืืฉืจ ื”ืื ืฉื™ื ืฉื ื‘ื—ืจื• ื‘ืืงืจืื™.
08:19
And you can get an intuitive appreciation for this
207
499260
2000
ื•ืืชื ื™ื›ื•ืœื™ื ืœืงื‘ืœ ื”ืขืจื›ื” ืื™ื ื˜ื•ืื™ื˜ื™ื‘ื™ืช ืœื–ื”
08:21
if you imagine just the people at the perimeter of the network.
208
501260
3000
ืื ืชื“ืžื™ื™ื ื• ืจืง ืืช ื”ืื ืฉื™ื ืฉื ืžืฆืื™ื ื‘ืงืฆื•ื•ืช ื”ืจืฉืช.
08:24
If you pick this person,
209
504260
2000
ืื ืชื‘ื—ืจื• ื‘ืื“ื ื”ื–ื”,
08:26
the only friend they have to nominate is this person,
210
506260
3000
ื”ืื“ื ื”ื™ื—ื™ื“ ืฉื”ื ื™ื‘ื—ืจื• ื‘ื• ื‘ืชื•ืจ ื—ื‘ืจ ื”ื•ื ื”ืื“ื ื”ื–ื”,
08:29
who, by construction, must have at least two
211
509260
2000
ื•ืœื• ืœืคื—ื•ืช ืฉื ื™ื™ื
08:31
and typically more friends.
212
511260
2000
ื•ื‘ืื•ืคืŸ ื˜ื™ืคื•ืกื™ - ื™ื•ืชืจ ื—ื‘ืจื™ื.
08:33
And that happens at every peripheral node.
213
513260
2000
ื•ื–ื” ืงื•ืจื” ื‘ื›ืœ ืฆื•ืžืช ื”ื ืžืฆืืช ื‘ืงืฆื” ื”ืจืฉืช.
08:35
And in fact, it happens throughout the network as you move in,
214
515260
3000
ื•ืœืžืขืฉื”, ื–ื” ืงื•ืจื” ื‘ื›ืœ ื”ืจืฉืช, ื›ื›ืœ ืฉืžืชืงื“ืžื™ื ืคื ื™ืžื”.
08:38
everyone you pick, when they nominate a random --
215
518260
2000
ื›ืœ ืื—ื“ ืฉืชื‘ื—ืจื•, ื›ืืฉืจ ื”ื ืžืฆื™ืขื™ื ืžื•ืขืžื“ ืืงืจืื™..
08:40
when a random person nominates a friend of theirs,
216
520260
3000
ื›ืืฉืจ ืื“ื ืืงืจืื™ ืžืฆื™ืข ื—ื‘ืจ ืฉืœื•
08:43
you move closer to the center of the network.
217
523260
3000
ืื ื• ืžืชืงืจื‘ื™ื ืœื›ื™ื•ื•ืŸ ืžืจื›ื– ื”ืจืฉืช.
08:46
So, we thought we would exploit this idea
218
526260
3000
ืื–, ื—ืฉื‘ื ื• ืฉื ื•ื›ืœ ืœื”ืฉืชืžืฉ ื‘ืจืขื™ื•ืŸ ื–ื”
08:49
in order to study whether we could predict phenomena within networks.
219
529260
3000
ืขืœ ืžื ืช ืœืœืžื•ื“ ื”ืื ืื ื• ื™ื›ื•ืœื™ื ืœื—ื–ื•ืช ืชื•ืคืขื•ืช ื‘ืชื•ืš ืจืฉืชื•ืช.
08:52
Because now, with this idea
220
532260
2000
ืžืื—ืจ ื•ืขื›ืฉื™ื•, ืขื ื”ืจืขื™ื•ืŸ ื”ื–ื”
08:54
we can take a random sample of people,
221
534260
2000
ืื ื• ื™ื›ื•ืœื™ื ืœืงื—ืช ืงื‘ื•ืฆื” ืืงืจืื™ืช ืฉืœ ืื ืฉื™ื,
08:56
have them nominate their friends,
222
536260
2000
ืœื‘ืงืฉ ืžื”ื ืœื”ืฆื™ืข ืืช ื—ื‘ืจื™ื”ื
08:58
those friends would be more central,
223
538260
2000
ืื ืฉื™ื ืืœื• ื™ื”ื™ื• ื™ื•ืชืจ ืžืจื›ื–ื™ื™ื,
09:00
and we could do this without having to map the network.
224
540260
3000
ื•ื ื•ื›ืœ ืœืขืฉื•ืช ื–ืืช ืœืœื ื”ืฆื•ืจืš ืœืžืคื•ืช ืืช ื”ืจืฉืช ื›ื•ืœื”.
09:03
And we tested this idea with an outbreak of H1N1 flu
225
543260
3000
ื‘ื“ืงื ื• ืืช ื”ืจืขื™ื•ืŸ ื”ื–ื” ืขื ื”ืชืคืจืฆื•ืช ืฉืœ ืฉืคืขืช H1N1
09:06
at Harvard College
226
546260
2000
ื‘ืงื•ืœื’' ื”ืจื•ื•ืืจื“
09:08
in the fall and winter of 2009, just a few months ago.
227
548260
3000
ื‘ืกืชื™ื• ื•ื—ื•ืจืฃ ืฉืœ 2009, ืœืคื ื™ ืžืกืคืจ ื—ื•ื“ืฉื™ื ื‘ืœื‘ื“.
09:11
We took 1,300 randomly selected undergraduates,
228
551260
3000
ืœืงื—ื ื• 1300 ืกื˜ื•ื“ื ื˜ื™ื ืืงืจืื™ื™ื ืœืชื•ืืจ ืจืืฉื•ืŸ,
09:14
we had them nominate their friends,
229
554260
2000
ื‘ื™ืงืฉื ื• ืžื”ื ืœื”ืฆื™ืข ื—ื‘ืจ ืื—ื“
09:16
and we followed both the random students and their friends
230
556260
2000
ื•ืขืงื‘ื ื• ืื—ืจื™ื”ื ื•ืื—ืจื™ ื”ื—ื‘ืจื™ื ืฉื‘ื—ืจื•
09:18
daily in time
231
558260
2000
ื™ื•ื ืื—ืจ ื™ื•ื
09:20
to see whether or not they had the flu epidemic.
232
560260
3000
ืขืœ ืžื ืช ืœื‘ื“ื•ืง ื”ืื ื”ื ื ื“ื‘ืงื• ื‘ืžื’ืคืช ื”ืฉืคืขืช.
09:23
And we did this passively by looking at whether or not they'd gone to university health services.
233
563260
3000
ื•ืขืฉื™ื ื• ื–ืืช ื‘ืื•ืคืŸ ืคืืกื™ื‘ื™, ืขืœ ื™ื“ื™ ื‘ื“ื™ืงื” ื”ืื ื”ื ื”ืœื›ื• ืœืฉื™ืจื•ืชื™ ื”ืจืคื•ืื” ืฉืœ ื”ืื•ื ื™ื‘ืจืกื™ื˜ื”
09:26
And also, we had them [actively] email us a couple of times a week.
234
566260
3000
ื•ื‘ื ื•ืกืฃ, ื‘ื™ืงืฉื ื• ืžื”ื ืœืฉืœื•ื— ืœื ื• ื“ื•ื"ืœ ืคืขืžื™ื™ื ื‘ืฉื‘ื•ืข.
09:29
Exactly what we predicted happened.
235
569260
3000
ืงืจื” ื‘ื“ื™ื•ืง ืžื” ืฉื ื™ื‘ืื ื•.
09:32
So the random group is in the red line.
236
572260
3000
ืื– ื”ืงื‘ื•ืฆื” ื”ืืงืจืื™ืช ืžืกื•ืžื ืช ื‘ืงื• ืื“ื•ื.
09:35
The epidemic in the friends group has shifted to the left, over here.
237
575260
3000
ื”ืžื’ืคื” ื‘ืงื‘ื•ืฆืช ื”ื—ื‘ืจื™ื- ื‘ื”ื™ืกื˜ ืœืฉืžืืœ, ื›ืืŸ.
09:38
And the difference in the two is 16 days.
238
578260
3000
ื•ื”ื”ื‘ื“ืœ ื‘ื™ืŸ ื”ืฉืชื™ื™ื ื”ื•ื 16 ื™ืžื™ื.
09:41
By monitoring the friends group,
239
581260
2000
ืขืœ ื™ื“ื™ ื ื™ื˜ื•ืจ ืงื‘ื•ืฆืช ื”ื—ื‘ืจื™ื,
09:43
we could get 16 days advance warning
240
583260
2000
ื”ื™ื™ื ื• ืžืกื•ื’ืœื™ื ืœืงื‘ืœ ื”ืชืจืื” ืžืจืืฉ ืฉืœ 16 ื™ืžื™ื
09:45
of an impending epidemic in this human population.
241
585260
3000
ืœืžื’ืคื” ืฉื‘ื“ืจืš ื‘ืื•ื›ืœื•ืกื™ื™ื” ื”ืื ื•ืฉื™ืช ื”ื–ื•.
09:48
Now, in addition to that,
242
588260
2000
ื•ื‘ื ื•ืกืฃ ืœื–ื”,
09:50
if you were an analyst who was trying to study an epidemic
243
590260
3000
ืื ื”ื™ื™ืช ืื ืœื™ืกื˜ ืฉืžื ืกื” ืœื—ืงื•ืจ ืžื’ืคื”
09:53
or to predict the adoption of a product, for example,
244
593260
3000
ืื• ืœื—ื–ื•ืช ืืช ื”ืื™ืžื•ืฅ ืฉืœ ืžื•ืฆืจ, ืœื“ื•ื’ืžื”,
09:56
what you could do is you could pick a random sample of the population,
245
596260
3000
ืžื” ืฉื”ื™ื™ืชื ื™ื›ื•ืœื™ื ืœืขืฉื•ืช ื”ื•ื ืœื‘ื—ื•ืจ ืžื“ื’ื ืืงืจืื™ ืฉืœ ื”ืื•ื›ืœื•ืกื™ื”,
09:59
also have them nominate their friends and follow the friends
246
599260
3000
ืฉื™ื‘ื—ืจื• ืžื™ื”ื ื—ื‘ืจื™ื”ื ื•ืœืขืงื•ื‘ ืื—ืจื™ ื”ื—ื‘ืจื™ื,
10:02
and follow both the randoms and the friends.
247
602260
3000
ื•ืœืขืงื•ื‘ ืื—ืจื™ ื”ืžื“ื’ื ื”ืืงืจืื™ ื•ืื—ืจื™ ื”ื—ื‘ืจื™ื.
10:05
Among the friends, the first evidence you saw of a blip above zero
248
605260
3000
ื‘ืงืจื‘ ื”ื—ื‘ืจื™ื, ื”ืจืื™ื” ื”ืจืืฉื•ื ื” ืฉืจืื™ืชื ืฉืœ ื ืงื•ื“ื” ืžืขืœ ื”ืืคืก
10:08
in adoption of the innovation, for example,
249
608260
3000
ื‘ืื™ืžื•ืฅ ืฉืœ ื”ื”ืžืฆืื”, ืœื“ื•ื’ืžื”,
10:11
would be evidence of an impending epidemic.
250
611260
2000
ื™ื”ื•ื•ื” ืจืื™ื” ืœืžื’ื™ืคื” ืžืžืฉืžืฉืช ื•ื‘ืื”.
10:13
Or you could see the first time the two curves diverged,
251
613260
3000
ืื• ืฉืชื•ื›ืœื• ืœืจืื•ืช ืืช ื”ืคืขื ื”ืจืืฉื•ื ื” ื‘ื” ืฉืชื™ ืขืงื•ืžื•ืช ืžืชืคืฆืœื•ืช,
10:16
as shown on the left.
252
616260
2000
ื›ืคื™ ืฉืจื•ืื™ื ืžืฉืžืืœ.
10:18
When did the randoms -- when did the friends take off
253
618260
3000
ืžืชื™ ื”ืืงืจืื™ื™ื...ืžืชื™ ื”ื—ื‘ืจื™ื ื”ืžืจื™ืื•
10:21
and leave the randoms,
254
621260
2000
ื•ื”ืฉืื™ืจื• ืืช ื”ืืงืจืื™ื™ื ืžืื—ื•ืจ,
10:23
and [when did] their curve start shifting?
255
623260
2000
ื•[ืžืชื™] ื”ืขืงื•ืžื” ืฉืœื”ื ื”ื—ืœื” ืœื–ื•ื–?
10:25
And that, as indicated by the white line,
256
625260
2000
ื•ื–ื”, ื›ืคื™ ืฉื”ืจืื” ื”ืงื• ื”ืœื‘ืŸ,
10:27
occurred 46 days
257
627260
2000
ืงืจื” 46 ื™ืžื™ื
10:29
before the peak of the epidemic.
258
629260
2000
ืœืคื ื™ ืฉื™ื ื”ืžื—ืœื”.
10:31
So this would be a technique
259
631260
2000
ื›ืš ืฉื–ื• ื™ื›ื•ืœื” ืœื”ื™ื•ืช ื˜ื›ื ื™ืงื”
10:33
whereby we could get more than a month-and-a-half warning
260
633260
2000
ืฉื‘ืขื–ืจืชื ื ื•ื›ืœ ืœืงื‘ืœ ื”ืชืจืื” ืฉืœ ื™ื•ืชืจ ืžื—ื•ื“ืฉ ื•ื—ืฆื™
10:35
about a flu epidemic in a particular population.
261
635260
3000
ืขืœ ืžื’ื™ืคืช ืฉืคืขืช ื‘ืื•ื›ืœื•ืกื™ื” ืžืกื•ื™ืžืช.
10:38
I should say that
262
638260
2000
ืขืœื™ ืœื•ืžืจ
10:40
how far advanced a notice one might get about something
263
640260
2000
ืฉื”ื”ืชืจืื” ืฉืื“ื ืขืฉื•ื™ ืœืงื‘ืœ ืœื’ื‘ื™ ืžืฉื”ื•
10:42
depends on a host of factors.
264
642260
2000
ืชืœื•ื™ื” ื‘ื”ืจื‘ื” ืžืื•ื“ ื’ื•ืจืžื™ื.
10:44
It could depend on the nature of the pathogen --
265
644260
2000
ื”ื™ื ื™ื›ื•ืœื” ืœื”ื™ื•ืช ืชืœื•ื™ื” ื‘ื˜ื‘ืขื• ืฉืœ ื”ืคืชื•ื’ืŸ --
10:46
different pathogens,
266
646260
2000
ืคืชื•ื’ื ื™ื ืฉื•ื ื™ื,
10:48
using this technique, you'd get different warning --
267
648260
2000
ื‘ืฉื™ืžื•ืฉ ื‘ื˜ื›ื ื™ืงื” ื–ื•, ืชืงื‘ืœื• ื”ืชืจืื•ืช ืฉื•ื ื•ืช --
10:50
or other phenomena that are spreading,
268
650260
2000
ืื• ืชื•ืคืขื•ืช ืžืชืคืฉื˜ื•ืช ืื—ืจื•ืช,
10:52
or frankly, on the structure of the human network.
269
652260
3000
ืื•, ื‘ื›ื ื•ืช, ืขืœ ืžื‘ื ื” ื”ืจืฉืช ื”ืื ื•ืฉื™ืช.
10:55
Now in our case, although it wasn't necessary,
270
655260
3000
ืขื›ืฉื™ื•, ื‘ืžืงืจื” ืฉืœื ื•, ืืฃ-ืขืœ-ืคื™ ืฉื–ื” ืœื ื”ื™ื” ื ื—ื•ืฅ
10:58
we could also actually map the network of the students.
271
658260
2000
ื™ื›ื•ืœื ื• ืœืžืขืฉื” ืœืžืคื•ืช ืืช ืจืฉืช ื”ืกื˜ื•ื“ื ื˜ื™ื.
11:00
So, this is a map of 714 students
272
660260
2000
ืื–, ื–ื• ืžืคื” ืฉืœ 714 ืกื˜ื•ื“ื ื˜ื™ื
11:02
and their friendship ties.
273
662260
2000
ื•ืงืฉืจื™ ื”ื—ื‘ืจื•ืช ืฉืœื”ื.
11:04
And in a minute now, I'm going to put this map into motion.
274
664260
2000
ื•ื‘ืขื•ื“ ื“ืงื” ืขื›ืฉื™ื•, ืื ื™ ืขื•ืžื“ ืœื”ื›ื ื™ืก ืืช ื”ืžืคื” ื–ื• ืœืชื ื•ืขื”.
11:06
We're going to take daily cuts through the network
275
666260
2000
ืื ื—ื ื• ื”ื•ืœื›ื™ื ืœืงื—ืช ืชืžื•ื ื•ืช ืžืฆื‘ ื™ื•ืžื™ื•ืช ื“ืจืš ื”ืจืฉืช,
11:08
for 120 days.
276
668260
2000
ื‘ืžืฉืš 120 ื™ืžื™ื.
11:10
The red dots are going to be cases of the flu,
277
670260
3000
ื”ื ืงื•ื“ื•ืช ื”ืื“ื•ืžื•ืช ืขื•ืžื“ื•ืช ืœื”ื™ื•ืช ืžืงืจื™ ืฉืคืขืช,
11:13
and the yellow dots are going to be friends of the people with the flu.
278
673260
3000
ื•ื”ื ืงื•ื“ื•ืช ื”ืฆื”ื•ื‘ื•ืช ืขื•ืžื“ื•ืช ืœื”ื™ื•ืช ื—ื‘ืจื™ื ืฉืœ ืื ืฉื™ื ืขื ืฉืคืขืช.
11:16
And the size of the dots is going to be proportional
279
676260
2000
ื•ื’ื•ื“ืœ ื”ื ืงื•ื“ื•ืช ืขื•ืžื“ ืœื”ื™ื•ืช ืคืจื•ืคื•ืจืฆื™ื•ื ืœื™
11:18
to how many of their friends have the flu.
280
678260
2000
ืœืžืกืคืจ ื—ื‘ืจื™ื”ื ื”ื—ื•ืœื™ื.
11:20
So bigger dots mean more of your friends have the flu.
281
680260
3000
ื›ืš ืฉื ืงื•ื“ื•ืช ื’ื“ื•ืœื•ืช ื™ื•ืชืจ ืื•ืžืจื•ืช ืฉืœื™ื•ืชืจ ืžื—ื‘ืจื™ืš ื™ืฉ ืฉืคืขืช.
11:23
And if you look at this image -- here we are now in September the 13th --
282
683260
3000
ื•ืื ืชืกืชื›ืœื• ื‘ืชืžื•ื ื” ื–ื• -- ื›ืืŸ ืื ื—ื ื• ืขื›ืฉื™ื• ื‘-13 ื‘ืกืคื˜ืžื‘ืจ --
11:26
you're going to see a few cases light up.
283
686260
2000
ืชื•ื›ืœื• ืœืจืื•ืช ื›ืžื” ืžืงืจื™ื ื ื“ืœืงื™ื.
11:28
You're going to see kind of blooming of the flu in the middle.
284
688260
2000
ืืชื ืขื•ืžื“ื™ื ืœืจืื•ืช ืกื•ื’ ืฉืœ ืฉื’ืฉื•ื’ ืฉืœ ื”ืฉืคืขืช ื‘ืžืจื›ื–.
11:30
Here we are on October the 19th.
285
690260
3000
ื›ืืŸ ืื ื• ื‘-19 ื‘ืื•ืงื˜ื•ื‘ืจ
11:33
The slope of the epidemic curve is approaching now, in November.
286
693260
2000
ื”ืฉื™ืคื•ืข ืฉืœ ืขืงื•ืžืช ื”ืžื’ื™ืคื” ืžืชืงืจื‘ ืขื›ืฉื™ื•, ื‘ื ื•ื‘ืžื‘ืจ.
11:35
Bang, bang, bang, bang, bang -- you're going to see lots of blooming in the middle,
287
695260
3000
ื‘ื•ื, ื‘ื•ื, ื‘ื•ื, ื‘ื•ื, ื‘ื•ื, ืืชื ืขื•ืžื“ื™ื ืœืจืื•ืช ื”ืจื‘ื” ืฉื’ืฉื•ื’ ื‘ืืžืฆืข,
11:38
and then you're going to see a sort of leveling off,
288
698260
2000
ื•ืื– ืืชื ื”ื•ืœื›ื™ื ืœืจืื•ืช ืกื•ื’ ืฉืœ ื”ืชื™ื™ืฉืจื•ืช,
11:40
fewer and fewer cases towards the end of December.
289
700260
3000
ืžืงืจื™ื ืžืขื˜ื™ื ื™ื•ืชืจ ื•ื™ื•ืชืจ ืœืงืจืืช ืกื•ืฃ ื“ืฆืžื‘ืจ.
11:43
And this type of a visualization
290
703260
2000
ื•ืกื•ื’ ื›ื–ื” ืฉืœ ื•ื™ื–ื•ืืœื™ื–ืฆื™ื”
11:45
can show that epidemics like this take root
291
705260
2000
ื™ื›ื•ืœ ืœื”ืจืื•ืช ืฉืžื’ื™ืคื•ืช ื›ืžื• ื–ื• ืžืฉืชืจืฉื•ืช ืืฆืœ
11:47
and affect central individuals first,
292
707260
2000
ื•ืžืฉืคื™ืขื•ืช ืขืœ ืคืจื˜ื™ื ืžืจื›ื–ื™ื™ื ืชื—ื™ืœื”,
11:49
before they affect others.
293
709260
2000
ืœืคื ื™ ืฉื”ืŸ ืžืฉืคื™ืขื•ืช ืขืœ ืื—ืจื™ื.
11:51
Now, as I've been suggesting,
294
711260
2000
ืขื›ืฉื™ื•, ื›ืคื™ ืฉื”ืฆืขืชื™,
11:53
this method is not restricted to germs,
295
713260
3000
ืฉื™ื˜ื” ื–ื• ืื™ื ื” ืžื•ื’ื‘ืœืช ืœื—ื™ื™ื“ืงื™ื,
11:56
but actually to anything that spreads in populations.
296
716260
2000
ืืœื ืœืžืขืฉื” ืœื›ืœ ื“ื‘ืจ ืฉืžืชืคืฉื˜ ื‘ืื•ื›ืœื•ืกื™ื•ืช.
11:58
Information spreads in populations,
297
718260
2000
ืžื™ื“ืข ืžืชืคืฉื˜ ื‘ืื•ื›ืœื•ืกื™ื•ืช.
12:00
norms can spread in populations,
298
720260
2000
ื ื•ืจืžื•ืช ื™ื›ื•ืœื•ืช ืœื”ืชืคืฉื˜ ื‘ืื•ื›ืœื•ืกื™ื•ืช.
12:02
behaviors can spread in populations.
299
722260
2000
ืฆื•ืจื•ืช ื”ืชื ื”ื’ื•ืช ื™ื›ื•ืœื•ืช ืœื”ืชืคืฉื˜ ื‘ืื•ื›ืœื•ืกื™ื•ืช.
12:04
And by behaviors, I can mean things like criminal behavior,
300
724260
3000
ื•ื‘ืฆื•ืจื•ืช ื”ืชื ื”ื’ื•ืช, ืื ื™ ื™ื›ื•ืœ ืœื”ืชื›ื•ื•ืŸ ืœื“ื‘ืจื™ื ื›ืžื• ื”ืชื ื”ื’ื•ืช ืคื•ืฉืขืช,
12:07
or voting behavior, or health care behavior,
301
727260
3000
ืื• ื”ืชื ื”ื’ื•ืช ืฉืœ ื”ืฆื‘ืขื”, ืื• ื”ืชื ื”ื’ื•ืช ืฉืงืฉื•ืจื” ื‘ื‘ืจื™ืื•ืช,
12:10
like smoking, or vaccination,
302
730260
2000
ื›ืžื• ืขื™ืฉื•ืŸ, ืื• ื—ื™ืกื•ืŸ,
12:12
or product adoption, or other kinds of behaviors
303
732260
2000
ืื• ืื™ืžื•ืฅ ืžื•ืฆืจื™ื, ืื• ืกื•ื’ื™ื ืื—ืจื™ื ืฉืœ ื”ืชื ื”ื’ื•ืช
12:14
that relate to interpersonal influence.
304
734260
2000
ื”ืงืฉื•ืจื™ื ืœื”ืฉืคืขื” ื‘ื™ืŸ-ืื™ืฉื™ืช.
12:16
If I'm likely to do something that affects others around me,
305
736260
3000
ืื ืกื‘ื™ืจ ืฉืืขืฉื” ืžืฉื”ื• ืฉืžืฉืคื™ืข ืขืœ ืื—ืจื™ื ืžืกื‘ื™ื‘ื™,
12:19
this technique can get early warning or early detection
306
739260
3000
ื˜ื›ื ื™ืงื” ื–ื• ื™ื›ื•ืœื” ืœืชืช ื”ืชืจืื” ืžื•ืงื“ืžืช, ืื• ืื™ืชื•ืจ ืžื•ืงื“ื,
12:22
about the adoption within the population.
307
742260
3000
ืขืœ ืจืžืช ื”ืื™ืžื•ืฅ ื‘ืื•ื›ืœื•ืกื™ื”.
12:25
The key thing is that for it to work,
308
745260
2000
ื”ื“ื‘ืจ ื”ื—ืฉื•ื‘ ื”ื•ื, ืขืœ ืžื ืช ืฉื–ื” ื™ืขื‘ื•ื“,
12:27
there has to be interpersonal influence.
309
747260
2000
ื—ื™ื™ื‘ืช ืœื”ื™ื•ืช ื”ืฉืคืขื” ื‘ื™ืŸ-ืื™ืฉื™ืช.
12:29
It cannot be because of some broadcast mechanism
310
749260
2000
ื–ื” ืœื ื™ื›ื•ืœ ืœืงืจื•ืช ื‘ื’ืœืœ ืกื•ื’ ื›ืœืฉื”ื• ืฉืœ ืžื ื’ื ื•ืŸ ืฉื™ื“ื•ืจ
12:31
affecting everyone uniformly.
311
751260
3000
ื”ืžืฉืคื™ืข ืขืœ ื›ื•ืœื ื‘ืื•ืคืŸ ืื—ื™ื“.
12:35
Now the same insights
312
755260
2000
ื›ืขืช ืื•ืชืŸ ื”ืชื•ื‘ื ื•ืช
12:37
can also be exploited -- with respect to networks --
313
757260
3000
ื™ื›ื•ืœื•ืช ืœื”ื™ื•ืช ืžื ื•ืฆืœื•ืช -- ื‘ื”ืชื™ื—ืก ืœืจืฉืชื•ืช --
12:40
can also be exploited in other ways,
314
760260
3000
ื™ื›ื•ืœื•ืช ื’ื ืœื”ื™ื•ืช ืžื ื•ืฆืœื•ืช ื‘ื“ืจื›ื™ื ืื—ืจื•ืช,
12:43
for example, in the use of targeting
315
763260
2000
ืœื“ื•ื’ืžื”, ื‘ืฉื™ืžื•ืฉ ื‘ื›ื™ื•ื•ืŸ ืืœ
12:45
specific people for interventions.
316
765260
2000
ืื ืฉื™ื ืกืคืฆื™ืคื™ื™ื ืœื”ืชืขืจื‘ื•ื™ื•ืช (interventions).
12:47
So, for example, most of you are probably familiar
317
767260
2000
ื›ืš, ืœืžืฉืœ, ืจื•ื‘ื›ื ืงืจื•ื‘ ืœื•ื•ื“ืื™ ืžื›ื™ืจื™ื
12:49
with the notion of herd immunity.
318
769260
2000
ืืช ืžื•ืฉื’ ื—ืกื™ื ื•ืช ื”ืขื“ืจ.
12:51
So, if we have a population of a thousand people,
319
771260
3000
ื›ืš, ืื ื™ืฉ ืœื ื• ืื•ื›ืœื•ืกื™ื” ืฉืœ ืืœืฃ ืื™ืฉ,
12:54
and we want to make the population immune to a pathogen,
320
774260
3000
ื•ืื ื• ืจื•ืฆื™ื ืœืขืฉื•ืช ืื•ื›ืœื•ืกื™ื” ื–ื• ื—ืกื™ื ื” ืœืคืชื•ื’ืŸ,
12:57
we don't have to immunize every single person.
321
777260
2000
ืื™ื ื ื• ื—ื™ื™ื‘ื™ื ืœื—ืกืŸ ื›ืœ ืื“ื ื•ืื“ื.
12:59
If we immunize 960 of them,
322
779260
2000
ืื ื ื—ืกืŸ 960 ืžื”ื
13:01
it's as if we had immunized a hundred [percent] of them.
323
781260
3000
ื›ืื™ืœื• ื—ื™ืกื ื• ืžืื” [ืื—ื•ื–ื™ื] ืžื”ื.
13:04
Because even if one or two of the non-immune people gets infected,
324
784260
3000
ืžื›ื™ื•ืŸ ืฉืืคื™ืœื• ืื ืื—ื“ ืื• ืฉื ื™ื™ื ืžื”ืื ืฉื™ื ื”ืœื ืžื—ื•ืกื ื™ื ื ื“ื‘ืง,
13:07
there's no one for them to infect.
325
787260
2000
ืื™ืŸ ืœื”ื ืื™ืฉ ืœื”ื“ื‘ื™ืง.
13:09
They are surrounded by immunized people.
326
789260
2000
ื”ื ืžื•ืงืคื™ื ื‘ืื ืฉื™ื ืžื—ื•ืกื ื™ื.
13:11
So 96 percent is as good as 100 percent.
327
791260
3000
ื›ืš ืฉ-96 ืื—ื•ื–ื™ื ื˜ื•ื‘ื™ื ื›ืžื• 100 ืื—ื•ื–ื™ื.
13:14
Well, some other scientists have estimated
328
794260
2000
ื•ื‘ื›ืŸ, ืžื“ืขื ื™ื ืžืกื•ื™ืžื™ื ืื—ืจื™ื ืืžื“ื•
13:16
what would happen if you took a 30 percent random sample
329
796260
2000
ืžื” ื™ืงืจื” ืื ืชื™ืงื—ื• ืžื“ื’ื ืืงืจืื™ ืฉืœ 30 ืื—ื•ื–ื™ื
13:18
of these 1000 people, 300 people and immunized them.
330
798260
3000
ืžืืœืฃ ื”ืื ืฉื™ื ื”ืืœื•, 300 ืื ืฉื™ื, ื•ืชื—ืกื ื• ืื•ืชื.
13:21
Would you get any population-level immunity?
331
801260
2000
ื”ืื ืชืงื‘ืœื• ื—ืกื™ื ื•ืช ื‘ืจืžืช ื”ืื•ื›ืœื•ืกื™ื”?
13:23
And the answer is no.
332
803260
3000
ื•ื”ืชืฉื•ื‘ื” ื”ื™ื ืœื.
13:26
But if you took this 30 percent, these 300 people
333
806260
2000
ืื‘ืœ ืื ื”ื™ื™ืชื ืœื•ืงื—ื™ื ืืช ืฉืœื•ืฉื™ื ื”ืื—ื•ื–ื™ื ื”ืืœื•, 300 ื”ืื ืฉื™ื ื”ืืœื”,
13:28
and had them nominate their friends
334
808260
2000
ื•ื”ื™ื™ืชื ืžื‘ืงืฉื™ื ืžื”ื ืœื”ืขืžื™ื“ ืœื‘ื—ื™ืจื” ืืช ื—ื‘ืจื™ื”ื
13:30
and took the same number of vaccine doses
335
810260
3000
ื•ื”ื™ื™ืชื ืœื•ืงื—ื™ื ืื•ืชื• ืžืกืคืจ ืฉืœ ืžื ื•ืช ื—ื™ืกื•ืŸ
13:33
and vaccinated the friends of the 300 --
336
813260
2000
ื•ื”ื™ื™ืชื ืžื—ืกื ื™ื ืืช ื—ื‘ืจื™ื”ื ืฉืœ ื”-300,
13:35
the 300 friends --
337
815260
2000
300 ื”ื—ื‘ืจื™ื,
13:37
you can get the same level of herd immunity
338
817260
2000
ืชื•ื›ืœื• ืœืงื‘ืœ ืื•ืชื” ืจืžื” ืฉืœ ื—ืกื™ื ื•ืช ืขื“ืจื™ืช
13:39
as if you had vaccinated 96 percent of the population
339
819260
3000
ื›ืื™ืœื• ื—ื™ืกื ืชื 96 ืื—ื•ื–ื™ื ืฉืœ ื”ืื•ื›ืœื•ืกื™ื”
13:42
at a much greater efficiency, with a strict budget constraint.
340
822260
3000
ื‘ื™ืขื™ืœื•ืช ื’ื“ื•ืœื” ื™ื•ืชืจ, ืชื—ืช ืžื’ื‘ืœื” ืชืงืฆื™ื‘ื™ืช ื ื•ืงืฉื”
13:45
And similar ideas can be used, for instance,
341
825260
2000
ื•ื ื™ืชืŸ ืœื”ืฉืชืžืฉ ื‘ืจืขื™ื•ื ื•ืช ื“ื•ืžื™ื, ืœื“ื•ื’ืžื”,
13:47
to target distribution of things like bed nets
342
827260
2000
ื›ื“ื™ ืœื›ื•ื•ืŸ ื”ืคืฆื” ืฉืœ ื“ื‘ืจื™ื ื›ืžื• ื›ื™ืœื•ืช
13:49
in the developing world.
343
829260
2000
ื‘ืขื•ืœื ื”ืžืชืคืชื—.
13:51
If we could understand the structure of networks in villages,
344
831260
3000
ืื ื ื•ื›ืœ ืœื”ื‘ื™ืŸ ืืช ืžื‘ื ื” ื”ืจืฉืชื•ืช ื‘ื›ืคืจื™ื,
13:54
we could target to whom to give the interventions
345
834260
2000
ื ื•ื›ืœ ืœื›ื•ื•ืŸ ืืฆืœ ืžื™ ื›ื“ืื™ ืœื”ืชืขืจื‘
13:56
to foster these kinds of spreads.
346
836260
2000
ืขืœ ืžื ืช ืœืขื•ื“ื“ ื”ืชืคืฉื˜ื•ื™ื•ืช ืžื”ืกื•ื’ื™ื ื”ืืœื”
13:58
Or, frankly, for advertising with all kinds of products.
347
838260
3000
ืื•, ื‘ื›ื ื•ืช, ืœืคืจืกื•ื ื›ืœ ืกื•ื’ื™ ื”ืžื•ืฆืจื™ื.
14:01
If we could understand how to target,
348
841260
2000
ืื ื ื•ื›ืœ ืœื”ื‘ื™ืŸ ื›ื™ืฆื“ ืœื›ื•ื•ืŸ,
14:03
it could affect the efficiency
349
843260
2000
ื–ื” ื™ื•ื›ืœ ืœื”ืฉืคื™ืข ืขืœ ื”ื™ืขื™ืœื•ืช
14:05
of what we're trying to achieve.
350
845260
2000
ืฉืœ ืžื” ืฉืื ื• ืžื ืกื™ื ืœื”ืฉื™ื’.
14:07
And in fact, we can use data
351
847260
2000
ื•ืœืžืขืฉื”, ืื ื• ื™ื›ื•ืœื™ื ืœื”ืฉืชืžืฉ ื‘ื ืชื•ื ื™ื
14:09
from all kinds of sources nowadays [to do this].
352
849260
2000
ืžืžื’ื•ื•ืŸ ืกื•ื’ื™ ืžืงื•ืจื•ืช ื›ื™ื•ื [ืขืœ ืžื ืช ืœืขืฉื•ืช ื–ืืช]
14:11
This is a map of eight million phone users
353
851260
2000
ื–ื•ื”ื™ ืžืคื” ืฉืœ ืฉืžื•ื ื” ืžื™ืœื™ื•ืŸ ืžืฉืชืžืฉื™ ื˜ืœืคื•ืŸ
14:13
in a European country.
354
853260
2000
ื‘ืืจืฅ ืื™ืจื•ืคื™ืช.
14:15
Every dot is a person, and every line represents
355
855260
2000
ื›ืœ ื ืงื•ื“ื” ื”ื™ื ืื“ื, ื•ื›ืœ ืงื• ืžื™ื™ืฆื’
14:17
a volume of calls between the people.
356
857260
2000
ืืช ื ืคื— ื”ืฉื™ื—ื•ืช ื‘ื™ืŸ ื”ืื ืฉื™ื.
14:19
And we can use such data, that's being passively obtained,
357
859260
3000
ื•ืื ื• ื™ื›ื•ืœื™ื ืœื”ืฉืชืžืฉ ื‘ื ืชื•ื ื™ื ื›ืืœื”, ืฉื”ื•ืฉื’ื• ื‘ืฆื•ืจื” ืคืกื™ื‘ื™ืช,
14:22
to map these whole countries
358
862260
2000
ื›ื“ื™ ืœืžืคื•ืช ืืจืฆื•ืช ืฉืœืžื•ืช
14:24
and understand who is located where within the network.
359
864260
3000
ื•ืœื”ื‘ื™ืŸ ืžื™ ืžืžื•ืงื ื”ื™ื›ืŸ ื‘ืชื•ืš ื”ืจืฉืช.
14:27
Without actually having to query them at all,
360
867260
2000
ื‘ืœื™ ืœืžืขืฉื” ืœืฉืื•ืœ ืื•ืชื ื›ืœืœ,
14:29
we can get this kind of a structural insight.
361
869260
2000
ื ื•ื›ืœ ืœืงื‘ืœ ืกื•ื’ ื›ื–ื” ืฉืœ ืชื•ื‘ื ื” ืžื‘ื ื™ืช.
14:31
And other sources of information, as you're no doubt aware
362
871260
3000
ื•ืžืงื•ืจื•ืช ืื—ืจื™ื ืฉืœ ืื™ื ืคื•ืจืžืฆื™ื”, ื›ืคื™ ืฉืืชื ืžื•ื“ืขื™ื ืœืœื ืกืคืง,
14:34
are available about such features, from email interactions,
363
874260
3000
ื–ืžื™ื ื™ื ื‘ื ื•ื’ืข ืœืžืืคื™ื™ื ื™ื ื›ืืœื”, ืžืื™ื ื˜ืจืืงืฆื™ื•ืช ื“ื•ื"ืœ,
14:37
online interactions,
364
877260
2000
ืื™ื ื˜ืจืืงืฆื™ื•ืช ืžืงื•ื•ื ื•ืช,
14:39
online social networks and so forth.
365
879260
3000
ืจืฉืชื•ืช ื—ื‘ืจืชื™ื•ืช ืžืงื•ื•ื ื•ืช, ื•ื›ืš ื”ืœืื”.
14:42
And in fact, we are in the era of what I would call
366
882260
2000
ื•ืœืžืขืฉื”, ืื ื—ื ื• ื‘ืชืงื•ืคื” ืฉืœ ืžื” ืฉืืงืจื
14:44
"massive-passive" data collection efforts.
367
884260
3000
ืžืืžืฆื™ ืื™ืกื•ืฃ ื ืชื•ื ื™ื "ืžืืกื™ื‘ื™ื™ื-ืคืกื™ื‘ื™ื™ื".
14:47
They're all kinds of ways we can use massively collected data
368
887260
3000
ืืœื• ื”ืŸ ืžื’ื•ื•ืŸ ื“ืจื›ื™ื ืฉื‘ื”ืŸ ืื ื—ื ื• ื™ื›ื•ืœื™ื ืœื”ืฉืชืžืฉ ื‘ื ืชื•ื ื™ื ืฉื ืืกืคื• ื‘ืฆื•ืจื” ืžืืกื™ื‘ื™ืช
14:50
to create sensor networks
369
890260
3000
ื›ื“ื™ ืœื™ืฆื•ืจ ืจืฉืชื•ืช ื—ื™ื™ืฉื ื™ื
14:53
to follow the population,
370
893260
2000
ืœืขืงื•ื‘ ืื—ืจื™ ื”ืื•ื›ืœื•ืกื™ื”,
14:55
understand what's happening in the population,
371
895260
2000
ืœื”ื‘ื™ืŸ ืžื” ืงื•ืจื” ื‘ืื•ื›ืœื•ืกื™ื”,
14:57
and intervene in the population for the better.
372
897260
3000
ื•ืœื”ืชืขืจื‘ ืœื˜ื•ื‘ื” ื‘ืื•ื›ืœื•ืกื™ื”.
15:00
Because these new technologies tell us
373
900260
2000
ืžื›ื™ื•ืŸ ืฉื”ื˜ื›ื ื•ืœื•ื’ื™ื•ืช ื”ื—ื“ืฉื•ืช ื”ืืœื• ืื•ืžืจื•ืช ืœื ื•
15:02
not just who is talking to whom,
374
902260
2000
ืœื ืจืง ืžื™ ืžื“ื‘ืจ ืขื ืžื™,
15:04
but where everyone is,
375
904260
2000
ืืœื ืื™ืคื” ื ืžืฆื ื›ืœ ืื—ื“,
15:06
and what they're thinking based on what they're uploading on the Internet,
376
906260
3000
ื•ืžื” ื”ื ื—ื•ืฉื‘ื™ื ื‘ื”ืชื‘ืกืก ืขืœ ืžื” ื”ื ืžืขืœื™ื ืœืื™ื ื˜ืจื ื˜,
15:09
and what they're buying based on their purchases.
377
909260
2000
ื•ืžื” ื”ื ืงื•ื ื™ื ื‘ื”ืชื‘ืกืก ืขืœ ืจื›ื™ืฉื•ืชื™ื”ื.
15:11
And all this administrative data can be pulled together
378
911260
3000
ื•ื›ืœ ื”ื ืชื•ื ื™ื ื”ืžื ื”ืœื™ื™ื ื”ืืœื• ื™ื›ื•ืœื™ื ืœื”ื™ืืกืฃ ื™ื—ื“
15:14
and processed to understand human behavior
379
914260
2000
ื•ืœื”ื™ื•ืช ืžืขื•ื‘ื“ื™ื ืขืœ ืžื ืช ืœื”ื‘ื™ืŸ ื”ืชื ื”ื’ื•ืช ืื ื•ืฉื™ืช
15:16
in a way we never could before.
380
916260
3000
ื‘ืื•ืคืŸ ืฉืžืขื•ืœื ืœื ื™ื›ื•ืœื ื• ืœืคื ื™ ื›ืŸ.
15:19
So, for example, we could use truckers' purchases of fuel.
381
919260
3000
ื›ืš, ืœื“ื•ื’ืžื”, ื ื•ื›ืœ ืœื”ืฉืชืžืฉ ื‘ืจื›ื™ืฉื•ืช ื”ื“ืœืง ืฉืœ ื ื”ื’ื™ ืžืฉืื™ื•ืช.
15:22
So the truckers are just going about their business,
382
922260
2000
ื”ื ื”ื’ื™ื ื”ืืœื• ืขื•ืกืงื™ื ื‘ืขื ื™ื™ื ื™ื”ื,
15:24
and they're buying fuel.
383
924260
2000
ื•ื”ื ืงื•ื ื™ื ื“ืœืง.
15:26
And we see a blip up in the truckers' purchases of fuel,
384
926260
3000
ื•ืื ื• ืจื•ืื™ื ืขืœื™ื” ื‘ืจื›ื™ืฉื•ืช ื”ื“ืœืง ืฉืœ ื ื”ื’ื™ ืžืฉืื™ื•ืช,
15:29
and we know that a recession is about to end.
385
929260
2000
ื•ืื ื• ื™ื•ื“ืขื™ื ืฉื”ืžื™ืชื•ืŸ ืขื•ืžื“ ืœื”ืกืชื™ื™ื.
15:31
Or we can monitor the velocity
386
931260
2000
ืื• ืื ื• ื™ื›ื•ืœื™ื ืœื ื˜ืจ ืืช ื”ืžื”ื™ืจื•ืช
15:33
with which people are moving with their phones on a highway,
387
933260
3000
ื‘ื” ืื ืฉื™ื ื ืขื™ื ืขื ื”ื˜ืœืคื•ื ื™ื ืฉืœื”ื ื‘ื›ื‘ื™ืฉ ืจืืฉื™,
15:36
and the phone company can see,
388
936260
2000
ื•ื—ื‘ืจืช ื”ื˜ืœืคื•ืŸ ื™ื›ื•ืœื” ืœืจืื•ืช,
15:38
as the velocity is slowing down,
389
938260
2000
ื›ืฉื”ืžื”ื™ืจื•ืช ืžืื˜ื”,
15:40
that there's a traffic jam.
390
940260
2000
ืฉื™ืฉ ืคืงืง ืชื ื•ืขื”.
15:42
And they can feed that information back to their subscribers,
391
942260
3000
ื•ื”ื ื™ื›ื•ืœื™ื ืœื”ื–ื™ืŸ ืืช ื”ืื™ื ืคื•ืจืžืฆื™ื” ื”ื–ื• ื‘ื—ื–ืจื” ืœืžื ื•ื™ื™ื ืฉืœื”ื,
15:45
but only to their subscribers on the same highway
392
945260
2000
ืื‘ืœ ืจืง ืœืžื ื•ื™ื™ื ืฉืœื”ื ื‘ืื•ืชื• ื›ื‘ื™ืฉ ืจืืฉื™
15:47
located behind the traffic jam!
393
947260
2000
ื”ืžืžื•ืงืžื™ื ืžืื—ื•ืจื™ ืคืงืง ื”ืชื ื•ืขื”!
15:49
Or we can monitor doctors prescribing behaviors, passively,
394
949260
3000
ืื• ืฉืื ื• ื™ื›ื•ืœื™ื ืœื ื˜ืจ ืืช ื”ืชื ื”ื’ื•ื™ื•ืช ืจื™ืฉื•ื ื”ืžืจืฉืžื™ื ืฉืœ ืจื•ืคืื™ื, ื‘ืื•ืคืŸ ืคืกื™ื‘ื™,
15:52
and see how the diffusion of innovation with pharmaceuticals
395
952260
3000
ื•ืœืจืื•ืช ื›ื™ืฆื“ ื”ืชืคืฉื˜ื•ืช ืฉืœ ื—ื™ื“ื•ืฉื™ื ืชืจื•ืคืชื™ื™ื
15:55
occurs within [networks of] doctors.
396
955260
2000
ืงื•ืจื™ืช ื‘ื™ืŸ [ืจืฉืช ืฉืœ] ืจื•ืคืื™ื.
15:57
Or again, we can monitor purchasing behavior in people
397
957260
2000
ืื• ืฉื•ื‘, ืื ื• ื™ื›ื•ืœื™ื ืœื ื˜ืจ ื”ืชื ื”ื’ื•ืช ืจื›ื™ืฉื” ืฉืœ ืื ืฉื™ื
15:59
and watch how these types of phenomena
398
959260
2000
ื•ืœืฆืคื•ืช ื›ื™ืฆื“ ืกื•ื’ื™ื ื›ืืœื” ืฉืœ ืชื•ืคืขื•ืช
16:01
can diffuse within human populations.
399
961260
3000
ื™ื›ื•ืœื™ื ืœื”ืชืคืฉื˜ ื‘ืื•ื›ืœื•ืกื™ื•ืช ืื ื•ืฉื™ื•ืช.
16:04
And there are three ways, I think,
400
964260
2000
ื•ื™ืฉ ืฉืœื•ืฉ ื“ืจื›ื™ื, ืื ื™ ื—ื•ืฉื‘,
16:06
that these massive-passive data can be used.
401
966260
2000
ืฉืœ ืฉื™ืžื•ืฉ ื‘ื ืชื•ื ื™ื ื”ืžืืกื™ื‘ื™ื™ื-ืคืืกื™ื‘ื™ื™ื ื”ืืœื”.
16:08
One is fully passive,
402
968260
2000
ืื—ืช ื”ื™ื ืคืืกื™ื‘ื™ืช ืœื—ืœื•ื˜ื™ืŸ,
16:10
like I just described --
403
970260
2000
ื›ืžื• ืฉืจืง ืชื™ืืจืชื™ --
16:12
as in, for instance, the trucker example,
404
972260
2000
ื›ืžื•, ืœื“ื•ื’ืžื”, ืืฆืœ ื ื”ื’ื™ ื”ืžืฉืื™ื•ืช,
16:14
where we don't actually intervene in the population in any way.
405
974260
2000
ืฉื ืื ื—ื ื• ืœื ืžืชืขืจื‘ื™ื ืœืžืขืฉื” ื‘ืื•ื›ืœื•ืกื™ื™ื” ื‘ื“ืจืš ื›ืœืฉื”ื™.
16:16
One is quasi-active,
406
976260
2000
ืื—ืช ื”ื™ื ืžืขื™ืŸ-ืืงื˜ื™ื‘ื™ืช,
16:18
like the flu example I gave,
407
978260
2000
ื›ืžื• ื‘ื“ื•ื’ืžืช ื”ืฉืคืขืช ืฉื ืชืชื™,
16:20
where we get some people to nominate their friends
408
980260
3000
ืฉื ืื ื• ืžื‘ืงืฉื™ื ืžืื ืฉื™ื ืžืกื•ื™ื™ืžื™ื ืœื”ืฆื™ืข ืืช ื—ื‘ืจื™ื”ื
16:23
and then passively monitor their friends --
409
983260
2000
ื•ืื– ืžื ื˜ืจื™ื ืืช ื”ื—ื‘ืจื™ื ื‘ืื•ืคืŸ ืคืกื™ื‘ื™ --
16:25
do they have the flu, or not? -- and then get warning.
410
985260
2000
ื”ืื ื”ื ื—ื•ืœื™ื ื‘ืฉืคืขืช, ืื• ืœื? -- ื•ืื– ืžืงื‘ืœื™ื ื”ืชืจืื”.
16:27
Or another example would be,
411
987260
2000
ืื• ื“ื•ื’ืžื” ืื—ืจืช ื™ื›ื•ืœื” ืœื”ื™ื•ืช,
16:29
if you're a phone company, you figure out who's central in the network
412
989260
3000
ืื ืืชื ื—ื‘ืจืช ื˜ืœืคื•ืŸ, ืืชื ืชื‘ื™ื ื• ืžื™ ืžืจื›ื–ื™ ื‘ืจืฉืช,
16:32
and you ask those people, "Look, will you just text us your fever every day?
413
992260
3000
ื•ืชื‘ืงืฉื• ืžื”ืื ืฉื™ื ื”ืืœื•, "ืชืจืื•, ื”ืื ืคืฉื•ื˜ ืชืฉืœื—ื• ืœื ื• ื˜ืงืกื˜ ืขื ื”ื—ื•ื ืฉืœื›ื ืžื“ื™ ื™ื•ื?
16:35
Just text us your temperature."
414
995260
2000
ืคืฉื•ื˜ ืชืฉืœื—ื• ืœื ื• ื˜ืงืกื˜ ืขื ื”ื˜ืžืคืจื˜ื•ืจื” ืฉืœื›ื."
16:37
And collect vast amounts of information about people's temperature,
415
997260
3000
ื•ืœืืกื•ืฃ ื›ืžื•ื™ื•ืช ืขืฆื•ืžื•ืช ืฉืœ ืื™ื ืคื•ืจืžืฆื™ื” ืขืœ ื”ื˜ืžืคืจื˜ื•ืจื” ืฉืœ ืื ืฉื™ื,
16:40
but from centrally located individuals.
416
1000260
2000
ืžืคืจื˜ื™ื ื”ืžืžื•ืงืžื™ื ื‘ืžืจื›ื–.
16:42
And be able, on a large scale,
417
1002260
2000
ื•ืœื”ื™ื•ืช ืžืกื•ื’ืœื™ื, ื‘ืงื ื” ืžื™ื“ื” ื’ื“ื•ืœ,
16:44
to monitor an impending epidemic
418
1004260
2000
ืœื ื˜ืจ ืžื’ื™ืคื” ืžืชืงืจื‘ืช
16:46
with very minimal input from people.
419
1006260
2000
ืขื ืงืœื˜ ืžื•ืขื˜ ืžืื•ื“ ืžืื ืฉื™ื.
16:48
Or, finally, it can be more fully active --
420
1008260
2000
ืื•, ืœื‘ืกื•ืฃ, ื–ื” ื™ื›ื•ืœ ืœื”ื™ื•ืช ืืงื˜ื™ื‘ื™ ื™ื•ืชืจ --
16:50
as I know subsequent speakers will also talk about today --
421
1010260
2000
ื›ืคื™ ืฉืื ื™ ื™ื•ื“ืข ืฉื’ื ื”ื“ื•ื‘ืจื™ื ื”ื‘ืื™ื ื™ื“ื‘ืจื• ืขืœ ื”ืืคืฉืจื•ืช ื”ื™ื•ื --
16:52
where people might globally participate in wikis,
422
1012260
2000
ืฉืื ืฉื™ื ืขืฉื•ื™ื™ื ืœื”ืฉืชืชืฃ ื’ืœื•ื‘ืœื™ืช ื‘ื•ื™ืงื™,
16:54
or photographing, or monitoring elections,
423
1014260
3000
ืื• ืฆื™ืœื•ื, ืื• ื ื™ื˜ื•ืจ ื‘ื—ื™ืจื•ืช,
16:57
and upload information in a way that allows us to pool
424
1017260
2000
ื•ืœื”ืขืœื•ืช ืืช ื”ืื™ื ืคื•ืจืžืฆื™ื” ื‘ื“ืจืš ืฉืชืจืฉื” ืœื ื• ืœืื’ื•ืจ
16:59
information in order to understand social processes
425
1019260
2000
ืื™ื ืคื•ืจืžืฆื™ื” ืขืœ ืžื ืช ืœื”ื‘ื™ืŸ ืชื”ืœื™ื›ื™ื ื—ื‘ืจืชื™ื™ื
17:01
and social phenomena.
426
1021260
2000
ื•ืชื•ืคืขื•ืช ื—ื‘ืจืชื™ื•ืช.
17:03
In fact, the availability of these data, I think,
427
1023260
2000
ืœืžืขืฉื”, ื–ืžื™ื ื•ืชื ืฉืœ ื”ื ืชื•ื ื™ื ื”ืืœื•, ืื ื™ ื—ื•ืฉื‘,
17:05
heralds a kind of new era
428
1025260
2000
ืžื›ืจื™ื–ื” ืขืœ ืกื•ื’ ืฉืœ ืชืงื•ืคื” ื—ื“ืฉื”
17:07
of what I and others would like to call
429
1027260
2000
ืฉืื ื™ ื•ืื—ืจื™ื ืจื•ืฆื™ื ืœืงืจื•ื ืœื”
17:09
"computational social science."
430
1029260
2000
"ืžื“ืข ื—ื‘ืจืชื™ ื—ื™ืฉื•ื‘ื™."
17:11
It's sort of like when Galileo invented -- or, didn't invent --
431
1031260
3000
ื–ื” ื‘ืขืจืš ื›ืžื• ื›ืืฉืจ ื’ืœื™ืœืื• ื”ืžืฆื™ื -- ืื•, ืœื ื”ืžืฆื™ื --
17:14
came to use a telescope
432
1034260
2000
ื”ืชื—ื™ืœ ืœื”ืฉืชืžืฉ ื‘ื˜ืœืกืงื•ืค
17:16
and could see the heavens in a new way,
433
1036260
2000
ื•ื”ื™ื” ื™ื›ื•ืœ ืœืจืื•ืช ืืช ื”ืจืงื™ืข ื‘ืฆื•ืจื” ื—ื“ืฉื”
17:18
or Leeuwenhoek became aware of the microscope --
434
1038260
2000
ืื• ื›ืืฉืจ ืœื™ื™ื‘ื ื”ื•ืง ื”ืคืš ืžื•ื“ืข ืœืžื™ืงืจื•ืกืงื•ืค --
17:20
or actually invented --
435
1040260
2000
ืื• ืœืžืขืฉื” ื”ืžืฆื™ื --
17:22
and could see biology in a new way.
436
1042260
2000
ื•ื”ื™ื” ืžืกื•ื’ืœ ืœืจืื•ืช ื‘ื™ื•ืœื•ื’ื™ื” ื‘ื“ืจืš ื—ื“ืฉื”.
17:24
But now we have access to these kinds of data
437
1044260
2000
ืืš ื›ืขืช ื™ืฉ ืœื ื• ื’ื™ืฉื” ืœืกื•ื’ื™ ื ืชื•ื ื™ื ื›ืืœื”
17:26
that allow us to understand social processes
438
1046260
2000
ืฉืžืืคืฉืจื™ื ืœื ื• ืœื”ื‘ื™ืŸ ืชื”ืœื™ื›ื™ื ื—ื‘ืจืชื™ื™ื
17:28
and social phenomena
439
1048260
2000
ื•ืชื•ืคืขื•ืช ื—ื‘ืจืชื™ื•ืช
17:30
in an entirely new way that was never before possible.
440
1050260
3000
ื‘ื“ืจืš ื—ื“ืฉื” ืœื—ืœื•ื˜ื™ืŸ ืฉืžืขื•ืœื ืœื ื”ื™ืชื” ืืคืฉืจื™ืช ืœืคื ื™ ื›ืŸ
17:33
And with this science, we can
441
1053260
2000
ื•ื‘ืขื–ืจืช ืžื“ืข ื–ื”, ืื ื• ื™ื›ื•ืœื™ื
17:35
understand how exactly
442
1055260
2000
ืœื”ื‘ื™ืŸ ื‘ื“ื™ื•ืง ื›ื™ืฆื“
17:37
the whole comes to be greater
443
1057260
2000
ื”ืฉืœื ื”ื•ืคืš ืœื”ื™ื•ืช ื’ื“ื•ืœ ื™ื•ืชืจ
17:39
than the sum of its parts.
444
1059260
2000
ืžืกื›ื•ื ื—ืœืงื™ื•.
17:41
And actually, we can use these insights
445
1061260
2000
ื•ืœืžืขืฉื”, ืื ื• ื™ื›ื•ืœื™ื ืœื”ืฉืชืžืฉ ื‘ืชื•ื‘ื ื•ืช ืืœื•
17:43
to improve society and improve human well-being.
446
1063260
3000
ืขืœ ืžื ืช ืœืฉืคืจ ืืช ื”ื—ื‘ืจื” ื•ืœืฉืคืจ ืืช ื”ืจื•ื•ื—ื” ื”ืื ื•ืฉื™ืช.
17:46
Thank you.
447
1066260
2000
ืชื•ื“ื” ืœื›ื
ืขืœ ืืชืจ ื–ื”

ืืชืจ ื–ื” ื™ืฆื™ื’ ื‘ืคื ื™ื›ื ืกืจื˜ื•ื ื™ YouTube ื”ืžื•ืขื™ืœื™ื ืœืœื™ืžื•ื“ ืื ื’ืœื™ืช. ืชื•ื›ืœื• ืœืจืื•ืช ืฉื™ืขื•ืจื™ ืื ื’ืœื™ืช ื”ืžื•ืขื‘ืจื™ื ืขืœ ื™ื“ื™ ืžื•ืจื™ื ืžื”ืฉื•ืจื” ื”ืจืืฉื•ื ื” ืžืจื—ื‘ื™ ื”ืขื•ืœื. ืœื—ืฅ ืคืขืžื™ื™ื ืขืœ ื”ื›ืชื•ื‘ื™ื•ืช ื‘ืื ื’ืœื™ืช ื”ืžื•ืฆื’ื•ืช ื‘ื›ืœ ื“ืฃ ื•ื™ื“ืื• ื›ื“ื™ ืœื”ืคืขื™ืœ ืืช ื”ืกืจื˜ื•ืŸ ืžืฉื. ื”ื›ืชื•ื‘ื™ื•ืช ื’ื•ืœืœื•ืช ื‘ืกื ื›ืจื•ืŸ ืขื ื”ืคืขืœืช ื”ื•ื•ื™ื“ืื•. ืื ื™ืฉ ืœืš ื”ืขืจื•ืช ืื• ื‘ืงืฉื•ืช, ืื ื ืฆื•ืจ ืื™ืชื ื• ืงืฉืจ ื‘ืืžืฆืขื•ืช ื˜ื•ืคืก ื™ืฆื™ืจืช ืงืฉืจ ื–ื”.

https://forms.gle/WvT1wiN1qDtmnspy7