A Machine Learning Approach for Intrusion Detection

Abstract

Abstract: Computer networks and virtual machine security are very essential in today’s era. IDS monitors a network or system for malicious action and protects a computer network from unofficial access from users, including perhaps insiders. Various existing systems have already been developed to detect malicious activity on target machines; sometimes any external user creates some malicious behavior and gets unauthorized access to victim machines to such a behavior system considered as malicious activities or Intruder. Machine Learning (ML) algorithms are applied in IDS in order to identify and classify security threats. Numerous machine learning and soft computing techniques are designed to detect the activities in real-time network log audit data. KKDDCUP99 and NLSKDD most utilized data sets to detect the Intruder on the benchmark data set. In this paper, we proposed the identification of impostors using machine learning algorithms. Two different techniques have been proposed a signature with detection and anomaly-based detection. The experimental analysis demonstrates SVM, Naïve Bayes, and ANN algorithms with various data sets and demonstrates system performance in the real-time network environment.