Misuse and Anomaly-Based Network Intrusion Detection System Using Fuzzy and Genetic Classification Algorithms

Intrusion Detection System (IDS) is a topic that has recently secured much interest in the computer security community. The main function of IDS is distinguishing and predicting normal or abnormal behaviors. The problem of intrusion detection has been studied and received a lot of attention in machine learning and data mining in the literature survey. The existing techniques are not effective to improve the classification accuracy and to reduce high false alarm rate. Therefore, it is necessary to propose new technique for IDS. In this paper, we propose a new Fuzzy C-Means clustering method and Genetic Algorithm for identifying intrusion and classification for both anomaly and misuse. The experiments of the proposed IDS are performed with KDD cup'99 data set. The experiments clearly show that the proposed method provides better classification accuracy over existing method.