Parts | Days | Selection | Search | Updates | Downloads | Help

SOE: Fachverband Physik sozio-ökonomischer Systeme

SOE 24: Group Dynamics

SOE 24.3: Talk

Thursday, March 14, 2013, 16:15–16:30, H37

Mixture models of human mobility — •Peter Reisenauer and Jörg Reichardt — University of Würzburg, Würzburg, Germany

Understanding human mobility patterns is key to many applications, ranging from e-commerce to traffic forecasting and containment of epidemics. We present a mixture model of Markov processes for human mobility, that allows for efficient estimation from data via maximum likelihood or Bayesian techniques. We discuss the possible benefits of this model for the prediction of movement of individuals, as well as theoretical limitations of the approach, using both real world and simulated data.