Detection and Analysis of Network Intrusions using Data Mining Approaches

摘要

The ultimate role of an intrusion detection system is to identify network threats or attacks in contrast to computing systems. The intrusion detection system (IDS) is one of the essential network protection device or software for guarding computing systems and it is proficient to identify and to examine network traffic data packets. Snort is free open-source software used as a network protection tool. Though, the Snort tool can detect only acknowledged attacks. In order to detect advanced network attacks, this research paper is developed based on advanced snort rules; k-SVM classification method is used for detection of network attacks. In this paper, the KDDCUP'99 dataset is used for the experimental study. The main goal of this research paper is to detect fraudulent network traffic. The main phases of research are data preparation, including the cleaning process, classification of the dataset, feature extraction, proposed snort rules, detection of attacks. The proposed system has produced effective detection rates.