Nicholas Christakis: How social networks predict epidemics

Nicholas Christakis: How social networks predict epidemics After mapping humans’ intricate social networks, Nicholas Christakis and colleague James Fowler began investigating how this information could better our lives. Now, he reveals…



quosmo1 says:

@fauxman omg! it totally does! 🙂 lol

sakar181 says:

@Timeofflux Substantiate your claim of quackery. Evidence, lets see it.

Scumbag-Bison says:

@AutodidacticPhd Firstly… “Each event in a given system is not unique”
“…they tend to share large number of common variables an behaviours”
Self-refuting implication. :/ “…the universe you describe would be pure
chaos, utterly incapable of supporting life” Our entire universe is a
chaotic system. “Mathematically, chaos refers to a very specific kind of
unpredictability: deterministic behaviour that is very sensitive to its
initial conditions.” I have sent you a PM with some information.

liesandtricks says:

its seems the only thing any of these resent video’s are about social media
and the environment (particularly to ocean), important topics of course but
there’s more to the world then that. what this looks like is another damm
tool for marketing

LudicrousTachyon says:

@Ambushcrysis It sounds like something the CIA or KGB would definately
abuse and I would never agree to do that. Maybe I’ll do it when there are
no governments that would care about picking people off. I’m sure your
method would produce greater accuracy, but it’s dangerous. While we all
imagine that technology will only be used for good, time and again it has
been shown to be used for evil. Perform an experiment with your friends.
Control group, his method, yours. See how much gain yours gives

morthim says:

@smudge6699 false, prove it. find five people between you and me.

asymmetrisch says:

@asymmetrisch Oh wait, my last sentence is bullshit, but the rest still

quosmo1 says:

this is such a load of crap

amjPeace says:

Kevin Bacon is at the very center!

AutodidacticPhd says:

@MarxIzalias Epidemics are prior state dependent complex systems that
progress through time, while lotteries are effectively instantaneous and
independent events (not even systems). The methods of analysis aren’t even
in the same branches of maths. Your comment on how little complexity there
is in a lottery should be your first clue that you have no basis for

darksean99 says:

@urcritic I see no reason why it can’t be used for both.

theron1n says:

A lot of the things he describes frighten me. Sure, you could use this
technique to monitor all types of behaviors, but should you?

thisisrobertb says:

So this is a chapter from The Tipping Point right?

Keisha Fabio says:

good thing not all people are affected by audio alone.. for some reason I
find this presentation with a somewhat negative vibe, perhaps his body
language, perhaps his choice of words/tone.. anyhow the idea is good &
potent. would be interesting if it is implemented in a good way. regards,
potentially paranoid person =]

steve0281 says:

@urcritic I’m more worried how governments will use it.

ra2monds says:

improve society? more like exploit society and sell more useless inventions!

urcritic says:

It is interesting… but I have a feeling this predictive strategy has more
of a chance of being adopted by corporations for marketing purposes rather
than by healthcare systems. Just call me an old fashioned cynic.

alphadawg44 says:

technology enabled tyranny

Ambushcrysis says:

@LudicrousTachyon If you’re already on a social networking site, then all
your friends are already on display for everyone to see (Unless you have it
hidden). So if someone created a program where all you had to do was go
through your friends list and check off who you know in real life, and that
information would automatically get sent to a database where it would then
be compared to other people’s friends, I’m sure the accuracy would be much
much more than 10%.

reafdaw01 says:

how is it that my friends are more central? If I am in the middle I will
nominate less central people..

WorldCollections says:

Gaytech !

flamablesteve says:

@urcritic Since when does cynicism fall out of fashion?

flamablesteve says:

@reafdaw01 But the majority of people picked will not be in the center, as
there are more people not in the center than in the center. There may be a
few people that end up nominating outward, but 99% of people will nominate
inward, making it almost entirely an increase.

jun lin says:

Leonard Hofstadter?

AutodidacticPhd says:

@Ambushcrysis “…and that information would automatically get sent to a
database where it would then be compared to…” Which would be a legal
headache to set up in the first place, and the instant it was leaked to the
press that someone was doing that, the site would die… practically
overnight. Besides, how well connected I feel to all the people on my
various friends lists changes almost daily. By the time the number
crunching was done the info would be out of date.

manilaenglish says:

Sounds like a bunch of hocus pokus, witch hunt bullshit to me.

AutodidacticPhd says:

@Ambushcrysis Even if some people explicitly agree to the study, you would
still have to get some form of agreement from the friends they elect. And
what you mean by “know irl” is not very rigorous… there are a lot of
people that I have known irl since before AOL was founded yet I never see,
and on the flip side there are people I know irl that don’t have a place on
any of my f-lists. For a simple academic study this would be no less work
and no better data than a more traditional method…

Scumbag-Bison says:

Continued; By working on a system of prediction, you undermine the system
before the outcome. There are circumstances this works but they are by far,
LESS COMPLEX and don’t have as many variables as social network constructs.
The results seem promising but the net result will be unjust if
preventative measures are actually taken, in a way it is presuming guilt by
association. If my friend likes X then I will. It is bunk.

Coreythecheese says:

@Tpar1234 Brilliantly said.

serioush says:

Great presentation, very interesting.

Barry Purcell says:

I think if it was actually true, it would be dangerous. However, the idea
that certain “node” people have undue influence over the entire population
is not new, and it’s been discredited: wiki/Small_world_experiment#Critiques

Eyal Lev says:

sounds like a good idea. say you want to know if/when a population might
turn towords being commuiniest, so you gather a few random ppl, ask them if
they are commuiniest, and than ask then to name a friend to be questioned.

WorldCollections says:

A bit persuaded…by the fake applause at the beguining.


How exactly is this supposed to be dangerous? Google already knows
everything from where you live, whom you relate yourself to, your interests
etc especially if you are using google chrome. but what evil are they gonna
use it for except maybe setting up ads according to your interests. Also I
find it funny how everyone is so anti-1984 when they give the most intimate
detail in their twitters and upload private pictures and videos for
everyone to see…

Ambushcrysis says:

@AutodidacticPhd people willingly providing that information for the sake
of the study, and how connected you felt to the person wouldn’t really
matter, what would matter the most is who you know irl.

redsbr says:

9:53 “or to predict the adoption of a product..” did he just say that? I
mean, why don’t you just start an epidemic WHILE predicting the adoption of
a vaccine youve also created for it.. you know…. since money and power is
the only thing corperations care about.

RonzigsGallery says:

Like everything that has ever been developed to improve the human
situation, there is a flip side which ALWAYS takes precedence. The power
structure we have allowed to evolve ensures that ALL new developments
intended to make things better will be co-opted by people and organizations
that will use them exclusively to control and subjugate the majority of
society to the will of the powerful elite.

roidroid says:

@amjPeace Kevin Bacon, in the Total Perspective Vortex.

Cory Y says:

hey, here’s a concept: when your friend has the flu, don’t be in close
proximity to them.

kommaV says:

Google has large enough data to apply this science!

brian cooper says:

The intimation you give away without even thinking about it. Let’s hope its
used for good

BlowDevilUp says:

I’m just not that interested in product adoption. Even if the product
advertisement has been crafted especially for me. In fact I would rather
isolate myself from any additional product advertisement altogether.

Julia Cotic says:

Lovely channel

samala51 says:

Data analytics are just amazing now that we have so much data at our
disposal .

hyperlexic2 says:

@urcritic why not both? but these models always break when we become
‘aware’ of them and apply them. ask the guys at Bear Stearns or Lehman Bros
how well their models predict markets.

kaosgoblin says:

Way to shout through your whole speech.

WoWanate says:

Bottom right, the girl at 4:51 looks fascinated.

BroBroDude says:

fuck this guy

Write a comment


This site uses Akismet to reduce spam. Learn how your comment data is processed.