Beyond the average: detecting global singular nodes from local features in complex networks

Abstract

Deviations from the average can provide valuable insights about the organization of natural systems. This article extends this important principle to the more systematic identification and analysis of singular local connectivity patterns in complex networks. Four measurements quantifying different and complementary features of the connectivity around each node are calculated and multivariate statistical methods are then applied in order to identify outliers. The potential of the presented concepts and methodology is illustrated with respect to a word association network.