Foreign Adversarial Activity and Influence Campaigns

Abstract

Influence campaigns are conducted in this environment by altering community and/or narrative. Content-based features and the network position of the message sender are key in predicting influence and the dispersion of the ideas. Despite the increase in findingsand technology in this area, there is still not a clear path for early detecting, assessing extent of impact, prioritizing responseefforts, and suggesting potential responses. To date mostly large scale attacks, or extreme attacks have been studied. Many suggest that all influence campaigns should be countered. Meanwhile, findings suggest that it might not be possible to counter extreme attacks head-on. In contrast, we find that influence campaigns take on a wide variety of forms, some of them are automatically countered by grass-root efforts, others naturally go away in a few days, and only a few types actually persist and require countering. The proposed research is to improve our ability to provide analysis and response at scale and across multiple languages that supports early identification, prioritization, and suggested counter-measures. A web-based analytic pipeline comprised of diverse high dimensional network techniques, machine learning techniques and natural language processing techniques will be used to analyze social media data from diverse platforms. The set of techniques will be varied to identify which type support early detection. A procedure forprioritizing influence campaigns will be developed. As part of the overall assessment process we will focus on diverse ongoing adversarial influence activities, counter-measure identification, and the development of guidelines for adversarial influence campaign assessment and associated counters based on level of prioritization and the nature of the threat.

Document Details

Document Type
DoD Grant Award
Publication Date
Jan 06, 2021
Source ID
N000142112105

Entities

People

  • Kathleen Carley

Organizations

  • Office of Naval Research
  • United States Navy

Tags

Readers

  • Computational Modeling and Simulation
  • Cybersecurity.
  • Political Violence and Terrorism Studies.

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference