Intrusion Detection is a serious global concern. The potential of network intrusion has posed a threat to national security; meanwhile the increasing prevalence of malware and incidents of network intrusions hinder the utilization of the Internet to its greatest benefit and incur significant economic losses to individuals, enterprises and public organizations.
In this thesis, an efficient algorithm for Intrusion Detection System as Tree-based Intrusion Detection System considering Data Features is proposed to enhance the misclassification, detection and false positive rate by considering data features.
Our results show a significant improvement in the misclassification, detection and false positive rate for the most difficult to detect attacks (e.g., Probing). In ours simulation, we used a Neural Network as classifier. This classifier basically shows lower performances than others. Nevertheless, our approach shows the better results in most cases. For that reason, if our approach, Tree-based Intrusion Detection System considering Data Features, is applied to other classifiers ( e.g., Support Vector Machine and Self-Organizing Map ) when design an intrusion detection system, we will get improved results.