Network Traffic Intrusion Detection by Aggregating Correlated Naive Bayes Predictions

A new intrusion classification scheme has been introduced which can effectively improve the classification performance when only few training data are available. In the proposed system, traffic flows are described using the statistical features and flow correlation information is modeled by IDS. Our proposed system is used to incorporate flow correlation information into the classification process. We introduced IDNB (Intrusion Detection using Naive Bayes)approach which enables the user to detect higher malicious behavior rates which does not greatly affect the network performances. NB is one of the classification methods which is an effective probabilistic classifier employing the Bayes’ theorem with naive feature independence assumptions applied in intrusion detection system. For each communication process, both the source and the destination are not malicious. NB classifier requires only a small amount of training data to estimate the parameters of a classification model.