Study from Microsoft Research in PNAS shows that when using public like data from…
Study from Microsoft Research in PNAS shows that when using public like data from Facebook you can build a whole profile about that person and can pretty accurately predict someone's race, religion and personality.
So be afraid off all these +Hello Kitty likers, they are instable 😉
http://goo.gl/AANnk
Private traits and attributes are predictable from digital records of human behavior
Abstract We show that easily accessible digital records of behavior, Facebook Likes, can be used to automatically and accurately predict a range of highly sensitive personal attributes including: sexu…
This post has been reshared 2 times on Google+
View this post on Google+
4 Replies to “Study from Microsoft Research in PNAS shows that when using public like data from…”
Have you looked at the supplementary table yet?
http://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1218772110/-/DCSupplemental/st01.pdf
Interestingly these are the 'Likes' highest correlated with Satisfaction with life (positively):
-Sarah Palin
-Glenn Beck
-Proud To Be Christian
-Indiana Jones
-Swimming
-Jesus Christ
-Bible
-Jesus
-Being Conservative
-Pride And Prejudice
Liking science was correlated with dissatisfaction. :/
Ah I didn't checked the table, only the other one with the colour plot about race and religion and stuff…. It's pretty scary how well some non-informative likes can turn into something informative 🙁
Surely by definition it's therefore informative?
Maybe not, and anyway, many of the likes they use are kind-off weird ones that I'd not really like that quickly. For example the religion one is a bit of a no-brainer that very christian people like the more christian president candidate….