Optimizing false alerts using multi-objective particle swarm optimization method

Security attacks through Internet have proliferated in recent years. An Intrusion Detection System in the general framework of a network catches the perpetrators before they do real damage to the resources. The performance of Intrusion Detection Systems usually depends on the false alarm rate and the detection rate, since true positive and false positive trade-off is always a major challenge in the choice of systems. In this paper, a multi-objective optimization approach using Particle Swarm Optimization method is used in the context of tradeoffs inherent in the performance evaluators of Intrusion Detection Systems.