SBIR/STTR Award attributes
This proposal for the Automatic Detection of Botnets and Cyborgs (ADBOC) system addresses the critical problem of detecting information campaigns by adversarial state and non-state actors early enough to help negate their schemes. Malicious state and non-state actors around the world increasingly use sophisticated online armies of bots and fake accounts in support of information campaigns that manipulate public opinion and influence world events. Our innovative ADBOC solution will automatically: (1) detect coordinated botnets and hybrid actors; (2) assess the role of these botnets within the overall information campaign and conflict space; and (3) present this information to human analysts in easily-understandable visualizations and dashboards. We are confident of our success in ADBOC design, implementation and transition because: (1) We have previously developed mature and proven algorithms for mapping social media based conflicts, and identifying behaviors described by conflict theory in real world data; (2) ADBOC builds upon our promising machine learning methods for identifying botnets; and (3) By applying two proven state-of-the-art social scientific methodologies – social role analysis and social influence analysis – we will illuminate a facet of the bot detection problem that is too often ignored by purely mathematically analyses that are not informed by social science.
