Network Discovery with Multi-intelligence Sources

was seeking a courier believed to be connected to Osama bin Laden. The courier was known as Abu Ahmed al-Kuwaiti, and his identity had been confirmed by more than one captured high-level al-Qaeda operative. In 2010, a telephone wiretap of a suspect led intelligence analysts to al-Kuwaiti. On the basis of this connection, al-Kuwaiti was eventually located in August 2010 and was tracked to a compound near Abottabad, Pakistan. After intelligence analysts gathered corroborating evidence, U.S. Navy SEALs raided the compound, and Osama bin Laden was found and killed. Working from tips and cues, such as that of al-Kuwaiti, and tracking them back through network connections created by communications, personal movement , and monetary transactions, intelligence analysts are able to uncover criminal and terrorist networks. By following the connections, rather than relying solely on the tips or cues, analysts can find higher-level network operatives, and network operations can be more completely disrupted. However, exploiting such connections requires resolving two related problems: network discovery and network exploration. The network-discovery problem consists of uncovering the subset of vertices within a network that exhibit a particular attribute or activity of interest. For example, given a social network, one may wish to detect all individuals belonging to a particular faction or gang. A related problem is network exploration. Often, the activities used to construct a graph (such as communications between computers or transit between locations) Analysts can glean much useful intelligence information from identifying relationships between individuals and groups, and tracking their activities. However, detecting networks of people and then investigating their activities are difficult tasks, especially in this era of information overload. Graph analysis has proven to be a useful tool for addressing these tasks, but it can be labor-intensive. To aid in this analysis, Lincoln Laboratory researchers developed a diffusion-based analytic that helps solve the problems of network discovery and prioritized exploration. This analytic, called threat propagation, has been demonstrated to effectively handle network detection and exploration tasks, and has been integrated into an interactive tool for generating networks from wide-area motion imagery. Multi-intelligence Graphs Mathematically, a graph G is simply a collection of vertices and edges, G = (V , E). The vertices represent the entities in the network, and the edges represent relationships between entities. In a multi-intelligence (multi-INT) graph, the entities encoded by the vertices are observable across a range of sensor modalities, and the relationships encoded in the …