IGIDS: Intelligent intrusion detection system using genetic algorithms

We present a genetic algorithm based network intrusion detection system named IGIDS, where the genetic algorithm is used for pruning best individuals in the rule set database. The process makes the decision faster as the search space of the resulting rule set is much compact when compared to the original data set. This makes IDS faster and intelligent. We generate possible intrusions which forms the basis for detecting intrusions on the network traffic. Our method exhibits a high detection rate with low false positives. We have used DARPA Dataset for initial training and testing purpose.