Survey of Random Forest Based Network Anomaly Detection Systems

Abstract: Network intrusion poses a serious threat to the security of financial and all other systems. The main objective of any online security system is to provide protection against malicious intentions of a user. The techniques used by intruders are bound to change and every day new methods of attacks on the network are being faced by all the systems on the net. One method to gain reliable security against unknown intrusions is to use Anomaly Detection Systems. Many existing intrusion detection systems are Rules Based, which have limitations when new intrusions appear. The proposed work intends to provide a system which detects network anomalies using machine learning (ML). The proposed system intends to improve the accuracy of anomaly detection as compared to the existing systems.

Keywords: Machine Learning, Intrusion Detection, Anomaly Detection.