Stochastic modeling of a terrorist event via the ASAM system

The ability of terrorist networks to conduct sophisticated and simultaneous attacks - the most recent one on March 11, 2004 in Madrid, Spain - suggests that there is a significant need for developing information technology tools for counter-terrorism analysis. These technologies could empower intelligence analysts to find information faster, share, and collaborate across agencies, "connect the dots" better, and conduct quicker and better analyses. One such technology, the adaptive safety analysis and monitoring (ASAM) system, is under development at the University of Connecticut. In this paper, the ASAM system is introduced and its capabilities are discussed. The vulnerabilities at the Athens 2004 Olympics are modeled and patterns of anomalous behavior are identified using a combination of feature-aided multiple target tracking, hidden Markov models (HMMs), and Bayesian networks (BNs). Functionality of the ASAM system is illustrated by way of application to two hypothetical models of terrorist activities at the Athens 2004 Olympics.