Thursday 25 Oct 2018: Core-periphery pairs in networks: algorithms and rationales
Dr. Naoki Masuda - University of Bristol
Harrison 170 14:30-15:30
With core-periphery structure of networks based on edge density, core nodes are densely interconnected, peripheral nodes are connected to core nodes to different extents, and peripheral nodes are sparsely interconnected. Core-periphery structure composed of a single core and a single periphery has been observed for various networks. We propose scalable algorithms to detect multiple non-overlapping groups of core-periphery structure in a network. We also show that core-periphery structure with the configuration model (a standard random graph model that perserves the (expected) degree of each node) as the null model is possible only when we allow at least three blocks of nodes. Therefore, conventional core-periphery structure composed of a single block of core nodes and a single block of peripheral nodes is denied, because such a structure would have only two blocks. We illustrate our algorithms with model and empirical networks.
Naoki Masuda received his PhD in 1998 from the University of Tokyo. He worked as Lecturer and then Associate Professor at the University of Tokyo between 2006 and 2014. He moved to University of Bristol, Department of Engineering Mathematics as Senior Lecturer, March 2014. His main research interests are network science and neuroimaging data analysis (using networks and statistical physics). His other research interests include evolutionary games, computational social science and collective behaviour.