Predicting Network Attacks Using Ontology-Driven Inference

Graph knowledge models and ontologies are very powerful modeling and reasoning tools. We propose an effective approach to model network attacks and attack prediction which plays important roles in security management. The goals of this study are: First we model network attacks, their prerequisites and consequences using knowledge representation methods in order to provide description logic reasoning and inference over attack domain concepts. And secondly, we propose an ontology-based system which predicts potential attacks using inference and observing information which provided by sensory inputs. We generate our ontology and evaluate corresponding methods using CAPEC, CWE, and CVE hierarchical datasets. Results from experiments show significant capability improvements comparing to traditional hierarchical and relational models. Proposed method also reduces false alarms and improves intrusion detection effectiveness.