An Intelligent Weighted Outlier Detection Method for Intrusion Detection Using Mst and K-nn

Intrusion Detection System (IDS) is a potential part in the area of network security system. An effective intrusion detection system is necessary for providing effective communications in the past world. The major challenging task in this system is the classification of users such as normal user and attacker. For that purpose so many classification algorithms have been proposed in the past to detect and report about user’s behaviour (normal or abnormal) in networks. The sufficient detection accuracy is not yet met due to the lacking of suitable methodology introduction in this field. Outlier detection is the effective process to improve the classification performance. In the past, many outlier detection methods with the combination of different clustering methods have been proposed. These all are have limitations in terms of accuracy and speed. In this paper, we propose a new outlier detection model called Intelligent Weighted MST and k-NN Based Outlier Detection (IWMKOD) to detect the intruders in all kinds of network environment. This model is the combination of the proposed Intelligent MST and kNN based Outlier Detection. [Xiao Chun Wang et al., 2015] and Weighted Distance based Outlier Detection [S. Ganapathy et al., 2011]. The experimental results show that the proposed algorithm improves the detection accuracy and reduces the false alarm rate.