Dr. Amy Sliva, Northeastern University – Artificial Intelligence and Cyber Security
In today’s Academic Minute, Dr. Amy Sliva of Northeastern University explains how advances in computing power are allowing for an artificially intelligent approach to cyber security.
Amy Sliva is an assistant professor at Northeastern University with a dual appointment in the Department of Political Science and the College of Computer and Information Science. Her research interests are focused on developing new artificial intelligence models of agent behavior to address problems in security policy, international conflict, and international development. She holds a Ph.D. from the University of Maryland, College Park.
Dr. Amy Sliva – Artificial Intelligence and Cyber Security
In today’s global landscape, conflict and uncertainty are everywhere. The problems we face in international security and development policy are difficult and complex, but my work aims to identify ways to better analyze and address them.
My research lies at the intersection of artificial intelligence and political science. It involves the development of artificial intelligence models of agent behavior to help interpret challenging issues and provide decision support for policy makers.
What I try to do is model, forecast and analyze behaviors of groups or organizations involved in global affairs—such as civil conflict, terrorism, and drug trafficking. I do this by examining a range of social, political, cultural, and economic factors that may impact decision-making.
My models of behavior use logic and probabilities to represent likely responses to certain situations—such as elections, protests, or attacks. Using a process called abductive reasoning, we work backwards to determine which policies are most likely to produce positive outcomes and prevent violence and terrorism.
For example, I have applied these artificial intelligence models to understanding the dynamics of the drug trade in Afghanistan and Pakistan and ethno political violence in the Middle East, specifically focusing on organizations such as Hezbollah and Hamas.
I recently started to apply these models to issues in cybersecurity where we want to understand the behaviors and strategies involved in cyber warfare, even though attribution—understanding who is actually perpetrating these activities—is very difficult and hard to understand.
The end goal of my work is to provide decision support for policy makers and analysts to help them identify better strategies and policies for conflict management, international development, or national security.