Exploiting Available Domain Knowledge to Improve Mining Aviation Safety and Network Security Data

This paper discusses a method for incorporating available domain knowledge into data mining techniques in order to improve the interestingness of the discovered rules. Existing domain knowledge is represented by a simple grammar and is used within the algorithms in order to reduce the search space and generate more interesting results. We implemented the proposed approach in the APriori and C4.5 algorithms and applied them to data from aviation safety and intrusion detection domains. Our experiments show promising results.