An intelligent intrusion detection system using average Manhattan distance-based decision tree

Abstract

Recently, security is an important challenge in Internet-based communication. In such a scenario, intrusion detection systems help to secure the information through the identification of normal and abnormal behaviors. In order to model these behaviors accurately and to improve the performance of the intrusion detection system, intelligent decision tree algorithm based on average Manhattan distance algorithm (IDTAMD) is proposed in this paper. In this proposed new classification algorithm for effective decision making in the network data set. Moreover, an attribute selection algorithm called modified heuristic greedy algorithm [1] is used to select itemsets from redundant data. The experimental results obtained in this work show high detection rates and reduce the false alarm rate. This system has been tested using the tenfold cross-validations on the KDD’99 Cup data set. The results have been tested with tenfold cross-validation.