POC: For inquiries regarding this work, please contact Dr. Robert Popp (rpopp@nsiteam.com).  

Description: Counter-Terror Social Network Analysis and Intent Recognition (CT-SNAIR) was an MIT Lincoln Laboratory program that developed automated tools for terror threat network analysis and tracking, intent recognition, and capability prediction. NSI developed techniques for stochastic scenario generation of terrorist attacks, as well as modeling of terrorist social networks and their intentions at the transactional level.

The scenerios were based on actual or realistic terrorist events (the 2005 Valentine’s Day bombings in the Philippines carried out by members of the Abu Sayyaf Group; the 1998 simultaneous IED bombings of USEmbassies in Kenya and Tanzania using explosives-packed trucks driven by suicide attackers; the 2009 al-Balawi suicide bombing of the CIA base at Forward Operating Base Chapman in Khowst, Afghanistan). NSI systematically generated threat scenarios from community-accepted scenario sources and open source literature, describing the scenarios in a narrative form with sufficient detailto support the TADL implementation; implementing multiple tests, experimental and demonstration scenarios in the TADL language; and designed and developed a standalone Attack Scenario Editor (ASE) in Java to support the creation of the TADL scenarios.

The modeling and simulation approach taken included probabilistic social network tracking on tagged, aggregated data sets across multiple relations; red team scenario modeling including a new terror attack description language (TADL); scenario-driven intent recognition via dynamic Bayesian networks; and use of external metadata and content-based metadata. NSI also explored modeling and simulation (M&S) and gaming tools for scenario and corpora library creation. Such scenarios are developed first in narrative form and subsequently in mathematical form to support the probabilistic simulation of the underlying events. NSI also designed and developed the advanced editor that supports the probabilistic coding of the scenarios. The editor has been designed and developed in a platform-independent manner, allowing implementation on a range of hardware and software platforms (e.g., Windows, Mac, Linux).