A Comparative Study on Various Intrusion Detection Techniques Using Machine Learning and Neural Network

Abstract

Security is of great concern these days as a large amount of data is being transmitted every hour in each domain. This paper provides an explained survey on the prevention and detection of intrusion using a neural network for the security of data. Various types of Intrusion Detection (IDS) and Intrusion Prevention System (IPS) have been studied and compared in this study. Moreover, Comparative study is made on different techniques like Network-Based, Host-Based, Anomaly Based, Signature-Based and Application-Based Intrusion Detection Techniques using Artificial Neural Network. The task of the IDS is to monitor the system for all sorts of malicious errors beforehand. The above techniques are compared based on the parameters like selecting the right feature, checking LAND attack, traffic filtering, blocking and many more. It can be concluded that based on the feature selection method Anomaly Detection is by far the best method for Intrusion Detection. This paper explains various models and techniques for an intrusion detection system. The paper highlights how can neural network and machine learning be used to detect Intruders using the IDS technique to prevent it from all malicious errors that can occur in the system.