Use of Genetic Algorithm with Fuzzy Class Association Rule Mining for Intrusion Detection

In today’s life Intrusion Detection System gain the attention, because of ability to detect the intrusion access efficiently and effectively as security is the major issue in networks. This system identifies attacks and reacts by generating alerts or blocking the unwanted data/traffic. Intrusion Detection System mainly classified as Anomaly based intrusion detection systems that have benefit of detecting novel attacks having false positive rate, and Misuse based intrusion detection systems fails to detect the novel attacks. The proposed system includes Genetic algorithm and the data mining method of fuzzy logic which is a class association rule mining. Genetic algorithm is used to extract the rules which are required for anomaly detection system. The use of the fuzzy logic in proposed model deals with mixed types of attribute and avoid sharp boundary problem. As association rule mining is used to extract the sufficient rules for the user purpose rather than to extract all the rules which are useful for the misuse detection. KDD dataset from the MIT Lincoln Laboratory is used which provides high detection rates. Keywords— Data Mining, Rule Mining, Intrusion Detection System (IDS), Genetic Algorithm (GA), Network Security, Fuzzy Logic.