INTRUSION DETECTION SYSTEM BASED ON FCA USING GA

Abstract

Intrusion detection is the process of supervising the events taking place in a computer system or network and analyzing them for signs of possible activities, which are assaulted or close to threats of violation for computer security. Incidents have many causes, such as malware (e.g., worms, spyware), attackers gaining unauthorized access to systems from the internet, and authorized users of systems who misuse their privileges or attempt to acquire extra privileges for which they are not empowered. Proposed method includes GA-based fuzzy Class Association Rule Mining with sub-attribute utilization and its application to classification, which can deal with discrete and continuous attributes at the same time. In addition, the proposed method is applied to both misuse detection and anomaly detection. Since the association rules used in the traditional information detection cannot effectively deal with alterations in network behaviour, it will better satisfy the real needs of abnormal detection to introduce the concept of fuzzy association rules to strengthen the adaptability. Experimental results with dataset KDD99Cup from MIT Lincoln Laboratory shows that the proposed method provides competitively high detection rate, i.e. 97% and PFR 3.45% compared with other machinelearning techniques.