Dresden 2017 – scientific programme
Parts | Days | Selection | Search | Updates | Downloads | Help
SOE: Fachverband Physik sozio-ökonomischer Systeme
SOE 6: Poster
SOE 6.12: Poster
Monday, March 20, 2017, 17:00–20:00, P2-OG4
Opinion formation on networks: the topology may predict more than we think. — •Michael Schnabel and Daniel Diermeier — University of Chicago, Chicago, USA
We consider opinion formation on a given network, using interacting spin systems on arbitrary network topologies as a modeling framework, and propose to project the high dimensional state space of possible opinion configurations onto two dimensions that quantify the average opinion as well as the average local consensus in the network. For an Ising model these macroscopic order parameters would correspond to the magnetization M and the energy E. Our approach, however, is not restricted to the Ising model. This two dimensional representation allows to obtain model independent insight into predominant opinion configurations that are imposed by the network topology alone. We devise a numerical method, based on the Wang-Landau algorithm, to determine the density of states for arbitrary network topologies which essentially sets the stage on which the flow of opinions can unfold once a particular dynamical model has been specified. Our framework may be used to draw general conclusions about the typical distribution of opinions in small to medium sized networks (up to O(103) nodes), e.g. such as identifying network topologies that would facilitate (or prevent) the occurrence of consensus or fragmentation of opinions in the network, and may be applicable for a broad range of dynamical models.