Anomaly-Based Intrusion Detection System: A modern Perspective

-Intrusion Detection Systems (IDS) are beneficial for providing security againstmalicious activities for the users. Attackers find new ways to send malicious packetswithout the detection of Traditional Signature Based IDS. Anomaly based intrusiondetection systems using Machine Learning are hence being researched for feasibility andthere are many models that have successfully classified the benchmark dataset NSL-KDDbut very few research papers have used UNB-IDS 2018. Hence, the approach used is tobuild a model that tries to successfully classify whether it is malicious or benign withlabelled packet information from both the datasets and compare them. Two datasets (Abenchmark dataset and a New and a Larger Dataset) are used to train the MachineLearning Classifiers and a sufficient testing accuracy of 90 percent and above withRandom Forests was achieved. The robustness of the model is tested against FGSM,JSMA and novel adversarial inputs using GANs. The model trained with NSL-KDDdataset showed significant decrease in prediction accuracy against adversarial inputswhile the Model trained with UNB-IDS Dataset is robust against adversarial inputs.