Is social physics based on big data a new theory of human interaction?

  • Level: Graduate 
  • Thread starter Thread starter microsansfil
  • Start date Start date
  • Tags Tags
    Physics
Join the discussion
Registration is free. Start your own thread to ask a follow-up.
Sign up Log in
8 replies · 5K views
microsansfil
Messages
325
Reaction score
43
Hello,

Is it a new theory of human social interaction, based on the concept of big data ?

http://socialphysics.media.mit.edu/

Patrick
 
Physics news on Phys.org

Discussion

No - but it does appear to claim to be a new specialist field in social sciences. Sort of.
The hope is that having lots of data will produce predictive models for human behavior.
Sounds like marketing and I see the links all seem to go to "media" dot MIT.
Is anyone doing it outside MIT?
 
Simon Bridge said:
Is anyone doing it outside MIT?
Sandy Pentland: "Social Physics: How Good Ideas Spread" seems to be the author of this notion

Patrick
 
Last edited by a moderator:
Pentland certainly seems to be the latest to popularize the term - he's with the "media" people group at MIT.
He's also done some work on social networks.

I don't think the field has a journal so far though.
Where does Sandy generally publish?
 
Simon Bridge said:
Where does Sandy generally publish?
Good question. I do not know this person. Even so, "Social Physics" (not social big data) seem to have a long history.

Patrick
 
The term has been used to mean different things through history.
I don't think the "human thermodynamics" site is a good source to use - it raises too many red flags for "crackpot".

Try:
http://en.wikipedia.org/wiki/Auguste_Comte
... August Comte used "socail physics" to mean the same thing as "sociology" ... the second word he used after the first was appropriated by others.
http://en.wikipedia.org/wiki/Adolphe_Quetelet
... for the related use to describe the use of statistical methods in social sciences.
Closer to the use by Pentland. However, the historical uses should be understood in context.
 
Simon Bridge said:
The term has been used to mean different things through history.
I don't think the "human thermodynamics" site is a good source to use - it raises too many red flags for "crackpot".
Ok

I do not know this site. I found it by chance. The content on this particular point "Social Physics" seemed to me credible.

I agree, always keep a critical reading. Especially when we have no knowledge on the subject

Simon Bridge said:
Try:
http://en.wikipedia.org/wiki/Auguste_Comte
... August Comte used "socail physics" to mean the same thing as "sociology" ... the second word he used after the first was appropriated by others.
http://en.wikipedia.org/wiki/Adolphe_Quetelet
... for the related use to describe the use of statistical methods in social sciences.
Closer to the use by Pentland. However, the historical uses should be understood in context.

Thank

Patrick
 
I guess even political scientists nowadays are more motivated to analyse their objects of study based on large data - it appears to be the newst trend in the field. However, this has still to show any real progress. What you mainly have in such fields are very simple models with virtually no predictive power. This happens partially because people in these fields did not receive the appropriate trainning to deal with mathematical modeling and large data sets, and partially because social behavior is affected by so many variables that it is hard to believe that anyone would have taken all of them into consideration.

I am pretty sure that you can find people working in this kind of approach in the Sociology, Antrophology, Economics and Political Science departments of big Universities.
 
It's mainly because it is easy to publish such theories in social science journals.
Prev. though there was a larger tendency to concentrate on qualitative research, so it's probably a move in the right direction (i.e. towards a harder form of science.)