Addressing strategic questions about policy implications, regional conflicts, or national interests typically requires an understanding and model of the behavior, goals, and intentions of collective actors, such as groups, organizations, and states. Typically, models and beliefs are tested against observed data to ensure their correspondence to reality. Unfortunately, data-driven methods of model testing and validation are at a disadvantage in strategic analysis because observed data is typically generated at a tactical or individual level. In other words, while the questions and models consider group, organization, or state-level actors, much of the observed data contains identifiers only of individual persons, devices, or events. In order to utilize the observed data in strategic analyses, the individual identifiers must be affiliated with the collective actors in the analysis.

The activities associated with this project will seek to understand and address the holistic affiliation problem. This encompasses both the problem of determining the groups, organizations, and/or states with which an individual entity affiliates themselves and the problem of utilizing these inferred affiliations to address questions about the strategic entities. Affiliation in this context can be taken to mean a variety of relationships, such as membership,association, or allegiance.

There are many applications of interest to LAS. Some examples include:

  • Privacy – Threats to privacy often arise from the chaining of information found in multiple data sources. Characterizing and mitigating these threats to privacy requires understanding the affiliations that could potentially be inferred from the integration of multiple data sources.
  • Radicalization – Terrorist and violently extremist groups achieve their goals through actions undertaken by people who affiliate themselves with the organizations. The LAS approach to the radicalization problem is to understand and anticipate the process by which individuals come to affiliate themselves with terrorist and violent extremist groups.
  • Attribution – Individual events are often indecipherable until they can be attributed to larger forces. Attribution is the process of affiliating entities and identifiers associated with an event to groups and organizations whose intentions and goals are driving the observed events.

Key areas of investigation for this project include but are not limited to:

  • Understanding and documenting analytic workflows for determining affiliation
  • Understanding how combining many sources of information could be used to increase confidence in affiliation assessment
  • Privacy-preserving methods of reasoning about group behavior using data about individuals
  • Tools and methods that support the development of scientifically rigorous analysis
  • Methods of reporting and consuming affiliation assessments
  • Methods of supporting affiliation analysis workflows through anticipatory analysis, social network analysis and/or large-scale computation
  • Enabling users to observe and control computational algorithms for affiliation
  • Methods of protecting information about affiliations
  • Methods of integrating models of group behavior with observations of individual behavior

Connect With Us

LAS aims to bring together a multi-disciplinary group of academic, industry, and government researchers, analysts and managers together to re-engineer the intelligence analysis process to promote predictive analysis. LAS will do this by conducting both classified and unclassified research in a variety of areas of research. The research done in this area will serve as the foundation for mission effects and integrated back into the enterprise.

If you would like to sign up for our email distribution list, please fill out this form: