Welcome to the FED-TL project repository! This project focuses on implementing Federated Learning (FL) with Transfer Learning (TL) to enhance network intrusion detection using the CSE-CICIDS 2018 dataset.

---

🚀 Project Overview

📖 What is FED-TL?

FED-TL combines Federated Learning and Transfer Learning to:

• Enable collaborative model training across distributed clients without sharing sensitive data.
• Improve detection accuracy for rare and zero-day attacks in network intrusion detection systems (NIDS).
• Reduce computation and communication costs by leveraging transfer learning.

---

🗂️ Repository Structure

Here’s what you’ll find in this repository:

• cic-ids_data_loader/: Contains preprocessed and distributed versions of the CSE-CICIDS 2018 dataset.
• CSE-CICIDS2018/: Implementation Python scripts for training, evaluation of the FL-TL models, including 4 clients adn 1 server.
• cic-ids_dataset.rar/: Contains all the files of CSE-CICIDS Dataset. Dataloader is in "cic-ids_data_loader/", it generates required datasets for clients and server.

---