Skip to Main Content
As traditional network intrusion detection based on pattern recognition can just get better classification accurately only with lots of prior knowledge which is difficult to obtain, a novel ensemble learning algorithm for fuzzy classification rules is presented. The fuzzy antecedent is adjusted based on the combination of ensemble learning and induction-enhanced particle swarm optimization for intrusion detection. By tuning the distribution of training instances and joining the distribution factor in computing of the fitness function, the collaboration of rules is taken into account during producing rules phase. So the classification error rate is reduced and the process of those rules in latter don't need. The results of experiments show that this method does not need plenty of prior knowledge about intrusion detection, can get good generalization and high classification rate.