Network AI - Malicious Traffic Analysis

This project aims to build an AI-powered system for capturing and analyzing network traffic to identify malicious activity in real-time. The tool leverages machine learning models and packet analysis to enhance network security.

---

Features

• Real-Time Packet Analysis: Capture and analyze live network traffic.
• Malware Detection: Use machine learning to classify traffic as "Normal" or "Malicious."
• Custom Dataset Support: Train models with your own labeled datasets for improved accuracy.
• User-Friendly Interface: A Tkinter-based GUI for easy interaction and visualization of results.

---

Installation

1. Clone this Repository:

git clone https://github.com/username/network_ai.git
cd network_ai

2. Create and Activate a Virtual Environment:

python -m venv .venv
source .venv/bin/activate  # For Windows: .venv\Scripts\activate

3. Install the Required Libraries:

pip install -r requirements.txt

---

How to Use

1. Launch the Application:

   python live_analyzer.py

2. Start Packet Sniffing:
   • Click "Start" to capture live network traffic.
   • Use "Browse" to analyze a pre-recorded .pcap file.
3. View Analysis Results:
   • The application classifies traffic and displays it in a structured table.
   • Use the "Details" section to inspect packet contents.
4. Custom Model Training:
   • Train your model with labeled datasets like UNSW-NB15 or CICIDS2017.
   • Replace the default model file (trained_model.pkl) with your trained model.

---

Notes

• Dataset Requirement: For accurate results, use well-labeled datasets. Suggested datasets include:
  • UNSW-NB15
  • CICIDS2017
• Custom Feature Extraction: Modify feature_extractor.py to include additional packet features as needed.

---

Requirements

• Python Version: 3.7 or higher
• Libraries:
  • Refer to the requirements.txt file for the complete list.

---

File Structure

.
├── live_analyzer.py      # Main application file with the GUI
├── feature_extractor.py  # Functions for extracting packet features
├── model_loader.py       # Handles loading the pre-trained machine learning model
├── packet_analysis.py    # Core packet analysis logic
├── train_model.py        # Script for training a custom model
├── requirements.txt      # Python dependencies
├── README.md             # Project documentation

---

License

This project is licensed under the MIT License. See the LICENSE file for details.

---

About the Project

This project was developed as part of a university course assignment. It was created entirely by me with the aim of building a network traffic analysis tool powered by machine learning. However, due to my limited knowledge and experience in this domain, I was unable to fully complete the project. While it is not fully functional, I hope it can serve as a learning resource for others or as a foundation for further development.