Network security assessment using a semantic reasoning and graph based approach

Abstract

Owing to the high value of business data, sophisticated cyber-attacks targeting enterprise networks have become more prominent, with attackers trying to penetrate deeper into and reach wider from the compromised machines. An important security requirement is that domain experts and network administrators have a common vocabulary to share security knowledge and quickly help each other respond to new threats. We propose an innovative ontology and graph-based approach for security assessment. An ontology is designed to represent security knowledge such as that of assets, vulnerabilities, and attacks in a common form. Using the inference abilities of the ontological model, an efficient system framework is proposed to generate attack graphs and assess network security. The performance of the proposed system is evaluated on test networks of differing sizes and topologies.