Scalable Heterogeneous Social Network Analysis

Driven by the nature of terrorist groups, the objective of this project is to develop robust and scalable algorithms and associated data structures for the analysis of complex social networks consisting of large numbers of individuals affiliated withmultiple, overlapping, structured organizations and linked by multiple types of interactions. Orion's Belt has already developed technology that successfully applies elements of social network analysis (SNA) to a database comprised of over 100,000 peopleand 30,000 organizations. In Phase I, we will extend this technology by developing new algorithms and data structures related to:. Management of hierarchical entities (multi-level organizations, morphing). Recognition of special nodes (bridgers and connectors). Computation of network metrics (connectivity, strength, density, centrality). Mechanisms for accommodating time (time-stamps for networks, changes over time)In Phase II, we will develop a full-functionality prototype implementing these new capabilities into our software package, and will add new functionality such as enhanced visualization. Greatly enhanced features and value for current and potentialcustomers. Improved effectiveness and efficiency for users. Significant advances in several key areas of social network analysis and other potential applications.