Cyberattack Detection

Machine learning based cyberattack detection in network traffic.

Context

• This works is part of the subject Security Analytics, 2020 Semester 2, The University of Melbourne.
• It was a personal assessment (no group project).

Files

• Please take a look to this report.

Main characteristics

The tasks of this project were:

• To apply unsupervised machine learning techniques for anomaly detection.
• to use gradient descent-based method to generate adversarial samples against supervised learning models

Structure

• This project works with the folder ml-data included in this delivery. In this folder, inside the subfolder 1_raw_data should be put the files necessary for the project.
• Inside the main cyberattack_detection folder there is several subfolder with the different files referenced in the report. This executables should work properly with the structure of ml-data.
• The requested output file are available in the folder 5 selected output

Requirements

• Environment for run a Jupyter Notebook. For example: Jupyter Project. A basic requirement for Jupyter Notebook is Python.
• The data files utilized in this project should be in a folder named 'data' in the same directory that the file 'Neural Network for movie genre prediction.ipynb'.

Compatibility

The libraries needed in this project are specified in the Jupyter Notebook. The most general libraries utilized are:

• Python: 3.8