Intrusion detection based on MinMax K-means clustering

Recently, with wide use of computer systems, internet, and rapid growth of computer networks, the problem of intrusion detection in network security has become an important issue of concern. In this regard, various intrusion detection systems have been developed for using misuse detection and anomaly detection methodologies. These systems try to improve detection rates of variation in attack types and reduce the false positive rate. In this paper, a new intrusion detection method has been introduced using MinMax K-means clustering algorithm, which overcomes the shortage of sensitivity to initial centers in K-means algorithm, and increases the quality of clustering. The experiments on the NSL-KDD data set indicate that the proposed method is more efficient than that based on K-means clustering algorithm. Also, the method has higher detection rate and lower false positive detection rate.