Despite the existence of various types of network intrusion detection system, growth of attacks at host level has increased in the present time. Therefore, there is a huge potential of research in this field and which motivates this research work. This paper analyses the pattern of four classes of attacks used to deploy host-based intrusion. KNN and Naïve-Bayes algorithms are employed and compared in this research work to determine the presence of intrusion using standard measures of performance.
