Cloud-based Network Intrusion Detection System using Deep Learning
Archana, H P Chaitra, Khushi Khushi, Pradhiksha Nandini, Sivaraman, Prasad B. Honnavalli
Abstract
With the increase in the internet traffic worldwide, it has been observed that there is a major spike in the influx of network traffic into any system over the years. The internet traffic and the number of attempts at malicious access to any organization have increased over the years. The attacker can take advantage of this and over flood the organization with dummy traffic and make the systems unresponsive. The current existing models of Network Intrusion Detection System (NIDS) have a good prediction rate but somehow have high False Positive Rate (FPR). We propose a Deep Learning (DL) model to reduce the FPR which offers a scalable solution by deploying the model on cloud to increase the responsiveness of the NIDS during high loads, hence increasing the availability. The model which is running on docker containers on the cloud instance can be accessed using REST APIs as it is deployed as microservice. The experimental results showed that Deep Neural Networks (DNN) with five hidden layers achieved the best accuracy of 95.02% and a least accuracy of 88.75% by Long Short-Term Memory (LSTM) with two layers. In traditional ML algorithms, Random Forest approach have achieved 86% accuracy.