A Data-driven, Decision Aid for Facility and Spatiotemporal Profiling (D3A-FSP)

The urban terrain shapes the strategy and tactics of both our adversaries and our military in the Global War on Terror. Hostile actors use man-made structures (facilities, buildings) for specific tasks and purposes (planning, supply, reconnaissance). The structures they use can be profiled in terms of their attributes (location, size, access, features, egress, utilities) and their utility (safe house, IED factory, observation post). Geoprofiling, a technique used in serial crime investigations, can also be applied to counter insurgency operations. However, like any analytical method, the benefit of geoprofiling is dependent on the specific application (i.e., criminal investigation), the availability of ‘good data’ to support the analysis, and the understanding of regional differences. Here, we propose to research and develop a data-driven, decision aid for facility and spatiotemporal profiling (D3A-FSP) that, given a set of adversary goals, finds the utility function from observed adversary behavior and generates inferences and predictions regarding the location of facilities that support adversary operations in the urban environment. D3A-FSP learns the utility function that the adversaries are using, and classifies and predicts the potential utility of a facility to the adversaries based on the derived metadata of each facility using belief networks.