Parts | Days | Selection | Search | Updates | Downloads | Help
SOE: Fachverband Physik sozio-ökonomischer Systeme
SOE 11: Networks: From Topology to Dynamics II (joint session SOE / DY / BP)
SOE 11.2: Talk
Tuesday, March 8, 2016, 15:15–15:30, H36
Geometric organization of real multiplex networks — •Kaj Kolja Kleineberg1, Marian Boguna1, M. Angeles Serrano1, and Fragkiskos Papadopoulos2 — 1Departament de Fisica Fonamental, Universitat de Barcelona, Marti i Franques 1, 08028 Barcelona, Spain — 2Department of Electrical Engineering, Computer Engineering and Informatics, Cyprus University of Technology
Real complex networks are organized to perform certain functions, among which targeted transport is important in a broad range of real systems, such as the Internet, social networks, or transportation networks. In reality, networks are not isolated entities but instead form interacting parts of larger and more complex systems. These systems are not a random combination of single networks, but instead are organized in a certain way. We investigate the geometric organization of multilayer networks and its implications. We find significant metric correlations between different layers. These correlations are key to answer many important questions concerning real multilayer systems. Metric correlations allow for inter-layer link prediction and the definition and detection of multidimensional communities. Metric correlations improve mutual greedy routing, that is targeted navigation in the whole multilayer system. Interestingly, only in the presence of metric correlations does the whole system outperform its single layers. We find that optimal correlations make multilayer systems perfectly navigable. Finally, we show how correlations present in the real Internet multiplex help navigating the digital world. Our findings have important implications for the design of real multilayer systems.