| Popis: |
Hidden networks arise in high-dimensional network structures when the hidden network members camouflage their existence by appearing randomly connected to the background network structure only, but in reality ensure they remain in persistent contact with one another over time. This paper analyses such networks through the novel use of group-based social network metrics to characterise the features of such networks when viewed from the micro-, meso and macro-level network perspetive. An analysis of the September 11 network, an extended September 11 network and a selection of popularly used simulated networks, will demonstrate the differences in hidden group detectability when the hidden group is embedded in a small network neigbourhood versus when the network neighbourhood is trebled in size. We will show that depending on the social metric used and the level of analysis, hidden groups can become more easily detectable with an increase in network size, particularly at the meso- and macro- levels, at which the detection of one hidden group will aid in unravelling the remainder of the September 11 network struture. Refereed/Peer-reviewed |
