Parts | Days | Selection | Search | Updates | Downloads | Help
DY: Fachverband Dynamik und Statistische Physik
DY 18: Networks II (with SOE)
DY 18.8: Talk
Wednesday, March 28, 2012, 12:15–12:30, H 0110
Networks of animal trade: from temporal paths to epidemic centrality — •Mario Konschake1,3, Hartmut Lentz1,2, Thomas Selhorst1, Igor M. Sokolov2, and Philipp Hövel3 — 1Friedrich-Loeffler-Institut, Wusterhausen — 2Humboldt-Universität, Berlin — 3Technische Universität, Berlin
Centrality of nodes is a major concept in network epidemiology which has been extensively studied. For temporal networks, however, a theoretical understanding is still in its infancy. In our case, a time-resolved topology arises from a series of static snapshots of the network taken at discrete time steps. The available animal-trade data has a temporal resolution of one day and includes 90.000 nodes.
We report on epidemiological relevant centrality measures based on the spread of an SIR-type disease. We analyse the robustness of the proposed measures under varying time of primary infection and under different infectious periods. We find the ranking of the nodes according to the measures sufficiently stable for different infectious periods, to be utilizable in the practical contexts of disease prevention and control. We thereby conclude that, for the analysed network, nodes with high epidemiological centrality can be identified independently of a specific disease.
We also find a threshold in the final size of the epidemic, so that diseases with an infectious period shorter than the intrinsic time constant of the network cannot propagate.