HYBRID MACHINE LEARNING TECHNIQUE FOR INTRUSION DETECTION SYSTEM

The utilization of the Internet has grown tremendously resulting in more critical data are being transmitted and handled online. Hence, these occurring changes have led to draw the conclusion that the number of attacks on the important information over the internet is increasing yearly. Intrusion is one of the main threat to the internet. Various techniques and approaches have been developed to address the limitations of intrusion detection system such as low accuracy, high false alarm rate, and time consuming. This research proposed a hybrid machine learning technique for network intrusion detection based on combination of K-means clustering and support vector machine classification. The aim of this research is to reduce the rate of false positive alarm, false negative alarm rate and to improve the detection rate. The NSL-KDD dataset has been used in the proposed technique. In order to improve classification performance, some steps have been taken on the dataset. The classification has been performed by using support vector machine. After training and testing the proposed hybrid machine learning technique, the results have shown that the proposed technique has achieved a positive detection rate and reduce the false alarm rate.