Detection of Attacks for IDS using Association Rule Mining Algorithm.

Intrusion detection system (IDS) plays a vital role in network infrastructure. Organizations have to protect the data from various attacks which are frequently affecting the networks. In this paper, we propose Association rule mining algorithm (ARMA) for detecting various network attacks such as smurf, neptune, mailbomb, back, apache2, processtable, guess_passwd, snmpguess, ipsweep, and nmap. KDD dataset contains three components, namely, “corrected dataset”, “10% dataset”, and “full dataset”, are employed for experimentation. Performances of the proposed ARMA are evaluated using the corrected dataset for training and other two datasets for testing. Java Development Kit (JDK) is used to conduct experiments and the results show significant improvement in the detection rate and also reduction of the false positive rate. [ABSTRACT FROM PUBLISHER]