The Defense Advanced Research Projects Agency (DARPA) has announced the researchers selected to its Influence Campaign Awareness and Sensemaking (INCAS) program. Research teams led by Smart Information Flow Technologies (SIFT), Protagonist Technology, University of Southern California Information Sciences Institute (ISI), University of Illinois Urbana-Champaign, and Uncharted Software will work to develop automated techniques and tools that aid U.S. analysts in the detection and sensemaking of geopolitical online influence campaigns. Additionally, teams led by Lockheed Martin Advanced Technology Laboratories and University of Maryland Applied Research Laboratory for Intelligence and Security (ARLIS) will support test, evaluation, and transition efforts of the resulting INCAS technologies.
“The U.S. is engaged with its adversaries in an asymmetric, continual war of weaponized influence narratives,” said Brian Kettler, the Information Innovation Office (I2O) program manager leading INCAS. “The combination of connectivity and modern media has greatly amplified the potential effects of geopolitical influence campaigns on populations. As adversaries seek to exploit misinformation as well as true information delivered via influence messaging, U.S. analysts need automated tools to help detect and make sense of adversarial influencer campaigns – allowing them to move away from largely manual and ad hoc approaches.”
Analysts currently use social listening tools – often developed for commercial purposes – to formulate complex keyword queries; track trending keywords, hashtags, and topics; and read hundreds to thousands of documents to identify influence themes. With these technologies, it’s difficult to connect messages over time and across multiple platforms to track how a campaign evolves. Further, rigorously assessing confidence in analytic conclusions is difficult, owing largely to the ad hoc, manual, subjective, and qualitative processes currently used.
INCAS aims to provide analysts with the ability to detect, characterize, and track geopolitical influence campaigns across multiple languages and platforms with confidence. The program plans to take on this technical research challenge across five research areas. The first is focused on the development of tools that enable analysts to directly and automatically detect implicit and explicit indicators of geopolitical influence in multilingual online messaging. Research teams working in this technical area will be led by SIFT, Protagonist Technology, and University of Southern California ISI.
To explain and anticipate population response to influence messaging, a second group of researchers will work to develop tools that dynamically segment the responding population and identify psychographic attributes relevant to geopolitical influence. Psychographic attributes, such as worldviews, morals, and values, are hypothesized to correlate more strongly with geopolitical influence response than the personality and demographic attributes used for marketing. As such, research teams led by University of Illinois Urbana-Champaign and University of Southern California ISI will work to develop techniques to segment the responding population, characterize each segment using psychographic and demographic attributes, and identify correlations among these attributes, influence indicators, and response.
The third research area will focus on the development of analyst-guided influence campaign modeling tools that accelerate analysts’ ability to detect and link influence indicators and messaging with population response over time and across multiple platforms. Researchers led by Uncharted Software will aim to create data modeling tools with rich human-machine interfaces (HMI) that allow analysts to assemble and compare campaign models and maintain confidence in the assessment.
Under the fourth research area, Lockheed Martin Advanced Technology Laboratories was selected to develop a testbed as well as provide social media messaging and other data feeds that the research teams can use to develop and evaluate their INCAS tools.
Finally, University of Maryland ARLIS will work to design and conduct technology evaluations, develop ground truth evaluation data for program scenarios, manage a multidisciplinary group of subject matter experts, and support transition evaluation efforts under the fifth research area.