Realtime Detection Of Network Anomalies Using Neural Network

Abstract

An Intrusion Detection System is model which monitors the network security from various type of Attacks. Intrusion Detection plays an important role in order to provide Network Security. In this paper we implement an Intrusion Detection System by building a Deep Learning Model using Feed Forward Neural Network (FFNN) and Long Short Term Memory Neural Network(LSTM). The study of model is based on Binary Classification and Multiclass Classification. The Model is implemented on Realtime Datasets or Dynamic Datasets. There is an comparative study between Feed Forward Neural Network and Long Short Term Memory Neural Network. The Intrusion Detection System(IDS) model improves the acccuracy and enlarge the further implementation for an Intrusion Detection Systems.