Networking and Cybersecurity

MODELING HUMAN BEHAVIOR FOR CYBERSECURITY AND SOCIAL SIMULATION

Human behavior is a key determining factor in assessing the effectiveness of an organization's cyber defenses, including it policies. Our current research aims to observe and model important aspects of human behavior in order to predict likely responses to security posture and the evolution of information in online social networks. Our cognitive agent architecture, DASH, captures important aspects of human online decision-making, including dual-process reasoning and mental models. 

Recently, in collaboration with the ISI's AI division, we have scaled DASH to handle distributed simulations with tens of millions of agents. This has been used to predict the behavior of individuals and groups in online settings such as Github, Twitter and Reddit. We also began an integration of our agent simulation with computational game theory in order to increase the range of problems to which it can be applied and the scale of games that can be solved. 

Additionally, we began an integration of our agent simulation with computational game theory in order to increase the range of problems to which it can be applied and the scale of games that can be solved.