SCAFFOLD – Synthesizing Course of Action Frames From Observations and Link Derivation

Predictive analytics tools, such as Insight, rely on example behavior models to feed their internal forecasting and inferencing components. These models represent potential courses of action (COAs) for targets of interest and are currently constructed by human experts—a costly and inefficient use of specialized human resources that bottlenecks the analytic process. To address this inefficiency, Stottler Henke proposes to develop the SCAFFOLD system to automatically construct the core “who/what/when/where” of these behavior models from reports of real events. First, SCAFFOLD will assemble COAs from observable events detected across diverse data sources. Second, it will correlate multi-source observations to “fill in the gaps” between temporally and geospatially referenced observations—even when represented differently among those sources. Finally, SCAFFOLD will infer implicit relationships to identify events that may influence an extracted COA. Construction of a proof-of-concept prototype during Phase I has answered research questions, reduced technical risk, and established a strong technological foundation for the continued development of the SCAFFOLD system. By applying the lessons learned and software developed during Phase I, we will demonstrate the full capabilities of SCAFFOLD by the end of Phase II, setting the stage to transition the technology into one or more civilian or DoD programs.