Translation of network metrics to behavior attributes Navy SBIR FY2009.1 Sol No.: Navy SBIR FY2009.1 Topic No.: N091-076 Topic Title: Translation of network metrics to behavior attributes Proposal No.: N091-076-0343 Firm: Infoscitex Corporation 303 Bear Hill Road Waltham, Massachusetts 02451-1016 Contact: Andrew DeCarlo Phone: (781) 890-1338 Web Site: www.infoscitex.com Abstract: Human social networks describe the social interactions among individuals and groups. The interactions of a human social network can be simple or complex, and the interactions among nodes can change depending on the social context. The ability to understand and accurately predict human behaviors has been sought for a long time, and recent studies of social networks by using network theory allow novel predictive models for human behaviors. However, current social network analysis approaches do not effectively identify the social interactions that reveal human behaviors, nor do they display the interactions among nodes or consider the effects of time on the social network. Infoscitex (IST) proposes the Reveal and Predict Social Behavior (RPSB) approach, which analyzes human social behavior with regard to social context. Additionally, RPSB uses directed social network graphs to show the direction of social interaction, and therefore hierarchy and relative importance within a network. RPSB also uses time delay and interaction time in considering an interaction's importance. At the completion of Phase II, we will integrate RPSB into a distributed software/firmware tool that conforms to service-oriented architecture (SOA) standards. Benefits: RPSB has several key benefits that will ensure its success in military, government, and commercial systems. RPSB analyzes interactions with respect to different events, ideas, and social contexts, allowing better understanding of how the events, ideas, and social contexts shape human interaction. Also, directed social network graphs show the relative importance of each individual in the network, and will help identify key individuals. RPSB uses time delay to show the likelihood of an interaction, and RPSB uses interaction time to show the importance of an interaction or series of interactions. RPSB has a very wide range of government, military, and commercial applications. The key military applications include providing intelligence agents with information about the structure of terrorist networks, and also modeling the effects of humans' reaction to events. RPSB can also be used by law enforcement to identify gang leaders and at-risk individuals and stop the spread of crime. Commercial applications are widely varied, and include personnel matching for employment networks, market research tools for advertising agencies, and personal shopping assistants. Return