Intrusion Detection with Tree-Based Data Mining Classification Techniques by Using Kdd Dataset

In the recent time a huge number of public and commercial services are used through via Internet, so that security of systems becomes most important issue in the society and threats from hackers also increased. So many researcher feels intrusion detection systems can be fundamental line of defense.  Intrusion Detection System (IDS) used against attacks for protected to the Computer networks. On another hand, data mining techniques can also contribute to intrusion detection. Intrusion detection can be classified into two classes: Anomaly based and Misuse based. One of the biggest problem with the anomaly base intrusion detection is detecting the number of high false alarm ratio.  In this paper solution will be provided to increase attack recognition rate with the minimum false alarm with the study of different tree-based data mining techniques. KDD cup dataset used for research purpose with WEKA tool.

Keywords: Data Mining; Intrusion Detection System; Decision Tree j48; Hoeffding Tree; Rep Tree; Random Forest; Random Tree; KDD dataset