Statistical analysis of different artificial intelligent techniques applied to Intrusion Detection System

Intrusion Detection System (IDS) which are increasingly a key part of system defense are used to identify abnormal activities in a computer system. In general, the traditional IDS relies on the extensive knowledge of security experts, in particular, on their familiarity with the computer system to be protected. To reduce this dependence, various data-mining and machine learning techniques have been used in the literature. The experiments and evaluations of the proposed intrusion detection system are performed with the NSL-KDD intrusion detection dataset. We will apply different learning algorithms on NSL-KDD data set, to recognize between normal and attack connections and compare their performing in different scenarios- discretization, features selections and algorithm method for classification- using a powerful statistical analysis: ANOVA. In this study, both the accuracy of the configuration of different system and methodologies used, and also the computational time and complexity of the methodologies are analyzed.