Visualization-Driven Statistical Analysis of Social Networks for Threat Detection and Characterization

• Activity Number: 417
• Type: Contributed
• Date/Time: Tuesday, August 3, 2010 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistics in Defense and National Security
• Author(s): Pranab K. Banerjee*+
• Companies: Space Dynamics Laboratory
• Keywords: Visualization ; Social ; Network ; Threat ; Analysis

Abstract

This research focuses on visualization driven statistical analysis of social networks. A suite of interactive as well as static visualization techniques have been developed for analyzing social networks for discovering potential adversarial entities. The analyses are based on statistical properties of social interaction profiles among individuals. The tools allow one to visually analyze a social network and discover organizational characteristics, such as hierarchy, inbound and outbound communication frequencies etc., that are typical in terrorist organizations. For concreteness, an openly available dataset consisting of cell phone call records from IEEE VAST 2008 challenge is used to show the effectiveness of the visualization tools. In this case, the interaction profiles include caller ID, recipient ID, call duration, caller location, time of call and other such cell phone records.