Knowledge Engineering in Cybersecurity

Dr. Sikos employs formal knowledge representation and ontology engineering in the Knowledge and Software Engineering Lab at the University of South Australia, and works closely with the DST Group and collaborates with Data61 and the Knowledge Discovery and Management Research Group of the University of Sydney. His current research projects are summarized below.

Network Knowledge Provenance

Proactive network monitoring, vulnerability assessment, and traffic path estimation rely on demanding processes that require expert knowledge and are time consuming, owing to the complexity of network topologies and network traffic flow. Therefore, the automated processing of network data is very much desired, however, constructing a machine-interpretable representation of network topologies and traffic flow is not trivial due to interoperability, complexity, and scalability issues. This project addresses many of these issues by utilizing Semantic Web standards, such as RDF and OWL, logically underpinned by description logics, which enable automated tasks to determine whether the traffic goes through a particular country, empower organizations to develop proactive cybersecurity policies, and inform decision makers in a timely manner.

Knowledge Representation

Modeling Complex Warfighting

In this project, recommendations are made on suitable modeling approaches for simulation in support of force design, and the corresponding inputs required by defense are identified. Operating concepts covered by the taxonomy under development include concepts of force-level electronic warfare, maritime force defense, space concepts, cooperative engagement capacity, and integrated air and missile defense.

Boeing EA-18G Growler

Project website

Further research interests of Dr. Sikos include, but are not limited to, video semantics and 3D model semantics, in particular content-based video retrieval and ontology-based video scene interpretation via spatiotemporal reasoning, and feature-based 3D model indexing and retrieval.