Analyzing Machine Learning Algorithms to Detect Novel attacks in Network Intrusion Detection System

Abstract

There are many Intrusion Detection System (IDS) based on signatures. If a new attacks, which by definition cannot be in the database of signatures is coming, those system cannot detect. The attackers will have a new way to bypass information system protection. There is just implementation of anomaly intrusion detection system, possible in theory to detect those new attacks. Many researchers try worked on the KDD 99 intrusion detection data set. There are many limitations of these researches such as high false positives and irrelevant alerts in detection of novel attacks (unknown attacks). In this proposed model, try to detect anomalies with high true positive rate and low false positive rate with the most efficient machine-learning algorithms based on decision trees and suggests an improvement to discover known and unknown attacks and show comparison results with these machine learning algorithms. Experimental results prove that the proposed method can get the high accuracy in detection those known and unknown attacks by using WEKA tool.