Description
Social network analysis focuses on modeling and understanding individuals of interest and their relationships. Aggregation of social networks can be used both to make analysis computationally easier on large networks, and to gain insight in subgroup interactions. Aggregation requires determining appropriate closely knit subgroups as well as choosing a measure or measures to represent the network data. This thesis provides the analyst with several techniques for using aggregation to analyze the characteristics of social networks. The contribution of this research lies in its ability to analyze a wide variety of social network structures and available data through two methods for subgroup detection and application of two network measures. These techniques are demonstrated first on notional social networks, then on open source information for the terrorist group, Jema’ah Islamiyah. Since analysts rarely have perfect information of the network structure, an exploration of the effects of missing arcs on subgroup detection is presented.
