A SVM and K-means Clustering based Fast and Efficient Intrusion Detection System

The intrusion or attack in the computer network is one of the most important issues creating problems for the network managers. However many countermeasures are taken for the security of the network but continuous growth of hackers requires to maintain the defending system up to data. This paper presents a K-means and support vector machine based intrusion detection system. The support vector machine is optimal partitioning based linear classifier and at least theoretically better other classifier also because only small numbers of classes required during classification SVM with one against one technique can be the best option and the K-means clustering filters the un-useful similar data points hence reduces the training time also hence provides an overall enhanced performance by reducing the training time while maintaining the accuracy. The proposed algorithm is tested using KDD99 dataset and results show the effectiveness of the algorithm. The paper also analyzed the effect of different input parameters on classification accuracy.