Homo Numericus Bis
Follow
Find tag "datascience"
1.3K views | +0 today
Homo Numericus Bis
humanités numériques
Curated by Mlik Sahib
Your new post is loading...
Your new post is loading...
Rescooped by Mlik Sahib from e-Xploration
Scoop.it!

Barabási : The Hidden #Pattern Behind Everything We Do I #datascience #SNA


Via ukituki, luiy
more...
ukituki's curator insight, November 24, 2013 8:35 AM

Barabasi argues that, rather than being random,
humans actually act in predictable patterns. We go along for long
periods of quiet routine followed suddenly by loud bursts of activity.
Barabasi demonstrates that these breaks in routine, or "bursts," are
present in all aspects of our existence— in the way we write emails,
spend our money, manage our health, form ideas. Barabasi has even
found "burstiness" in our webpage clicking activity and the online
news cycle."

Rescooped by Mlik Sahib from e-Xploration
Scoop.it!

'Social dispersion': the Facebook factor that predicts relationships - and when they will end | #datascience

'Social dispersion': the Facebook factor that predicts relationships - and when they will end | #datascience | Homo Numericus Bis | Scoop.it
A scientific paper authored by a computer scientist and a senior engineer at Facebook has shown how your online social networks not only reveal who you’re going out with, but also when you’ll break up (and yes, that's without checking your...

Via luiy
more...
luiy's curator insight, October 31, 2013 1:12 PM

Backstrom and Kleinberg found that looking at just the number of mutual friends between any two individuals – a factor known as ‘embeddedness’ - was not actually a strong indicator the pair were in a relationship, and that instead a quality known as ‘dispersion’ was far more telling.

 

Dispersion measures not only mutual friends but the network structures that connect these friends together. ‘Low dispersion’ – the quality that was associated with couples – indicates not only that two people have a large number of mutual friends, but also that these mutual friends knew one another.

 

Essentially, romantic partners act as social bridges between individuals’ networks, introducing people to each other and creating friendships. Eg, you might go for drinks with your boyfriend's friends from work and bring some of your friends from home to meet them.

Using this dispersion algorithm Backstrom and Kleinberg  were able to correctly identify who somebody’s spouse was 60 per cent of the time and correctly guess somebody’s partner a third of the time – a far better return than the 2 per cent success rate from pure guesswork.