Yoshida, Ruriko

Malware is malicious software that can be used to disrupt computer operations, gather sensitive information without consent, gain access to private computer systems, or display unwanted advertising, among other things. As modern, smart communities become more and more dependent on interconnected and intraconnected information systems, malware is becoming a bigger threat to our societal welfare and well-being. Hence, it is becoming increasingly important to analyze how malware codes are implemented and developed as a means to protect sensitive information in the cyber space. One of the mathematical constructs used to represent the evolution of malware codes is directed acyclic graphs (DAGs) or phylogenetic networks, which can be used to represent the relationships among different samples of malware code. In order to do the above, a proper model for graph generation needs to be selected. This process is typically referred to as a model selection. This proposed project aims to study centrality-based, Bayesian approaches to establishing the similarity level between a network and a model.
Operations Research
National Science Foundation
NSF
2017