J. Kertesz, Budapest University of Technology and Economics , Hungary

Title:
Temporal aspects of social behavior from mobile phone data

Abstract:
Digitalized personal data provide a gold mine for social science. For example, 
as mobile phones have become one of the main communication tools with an almost 
entire coverage of the adult population, the study of detailed prescription and 
call billing records enable insight into the structure and dynamics of the whole 
society. By analyzing such data we uncovered [1] the relationship between community 
structure and link intensities and proved Granovetter's hypothesis [2] on the 
"importance of weak ties": Strongly wired communities are linked together by weak ties. 
The society is known to be a small world, however, spreading of information, gossips 
and alike through one to one connections is a surprisingly slow process. We showed that 
this slowing down originates mostly in the mentioned "Granovetter-type" relationship 
between topology and link weights, as well as in the bursty character of human activities. 
In order to be able to identify the contributions of the different possible components in 
slowing down the spreading process, a series of null models were introduced with targeted 
shufflings of the original data. [3] The analysis of the time resolved call sequences shed 
light onto the relationship between the dynamic behavior of individuals and their topological 
role in the social network and, in general, onto the detailed dynamics of interpersonal communication [4].

[1] J.-P. Onnela et al., Proc. Natl. Acad. Sci. (USA) 104, 7332 (2007) and
J.-P. Onnela et al., New J. Phys. 9, 179 (2007)
[2] M. Granovetter, Am. J. Sociol. 78, 1360 (1973)
[3] M. Karsai et al. arXiv:1006.2125
[4] L. Kuvanen et al. (in preparation)