Hands-on Lab

Introduction

This lab focuses on creating an IDS that classifies network traffic as normal or anomalous using a custom neural network.

Objectives

- Understand how neural networks classify network traffic.

- Train and evaluate the IDS model.

- Visualize performance metrics.

Lab Steps

1. Dataset Preparation:

   - Use the network traffic dataset.

   - Preprocess features and labels (e.g., encode 'normal' and 'anomalous').

2. Train the Neural Network:

   - Design a classification network.

   - Train the model on the training data.

3. Evaluate the IDS:

   - Calculate accuracy, precision, recall, and F1-score.

   - Plot the confusion matrix and ROC curve.

4. Experiment with Adversarial Testing:

   - Use GAN-generated synthetic anomalies to evaluate IDS robustness.