BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Interdependence and Predictability of Human Mobility and Social In
 teractions - Mirco Musolesi (University of Birmingham)
DTSTART:20130521T140000Z
DTEND:20130521T150000Z
UID:TALK44907@talks.cam.ac.uk
CONTACT:Eiko Yoneki
DESCRIPTION:The study of the interdependence of human movement and social 
 ties of individuals is one of the most interesting research areas in compu
 tational social science. Previous studies have shown that human movement i
 s predictable to a certain extent at different geographic scales. One of t
 he open problems is how to improve the prediction exploiting additional av
 ailable information. In particular\, one of the key questions is how to ch
 aracterise and exploit the correlation between movements of friends and ac
 quaintances to increase the accuracy of the forecasting algorithms.\n\nIn 
 this talk I will discuss the results of our analysis of the Nokia Mobile D
 ata Challenge dataset showing that\, by means of multivariate nonlinear pr
 edictors\, it is possible to exploit mobility data of friends in order to 
 improve user movement forecasting. This can be seen as a process of discov
 ering correlation patterns in networks of linked social and geographic dat
 a. I will also show how mutual information can be used to quantify this co
 rrelation\; I will demonstrate how to use this quantity to select individu
 als with correlated mobility patterns in order to improve movement predict
 ion. Finally\, I will show how the exploitation of data related to friends
  improves dramatically the prediction with respect to the case of informat
 ion of people that do not have social ties with the user.\n\nMore informat
 ion about this project can be found at this URL : http://www.cs.bham.ac.uk
 /research/projects/nsl/mobility-prediction/\n
LOCATION:FW26\, Computer Laboratory\, William Gates Builiding
END:VEVENT
END:VCALENDAR