Predicting Adversarial Behaviours and Their Motivation for Automated Network Defense

The project designed a framework modelling the motivations, cognitive antecedents and dynamic decision-making processes of adversaries in the lead-up to, as well as during, a cyber-attack. Concurrently, novel algorithms for graph embedding with application to anomaly and attack detection were developed. The framework and tools developed in the research was validated using a Capture-the-Flag experiment to aid in further development of novel algorithms for predicting adversarial behaviors and automating network defense and intrusion detection.