Analysis of Network flows, different cyber attacks and Android Malware. University Project - Artificial Intelligence for Security
Welcome to our project.
Our group of 4 Cyber Risk Strategy and Governance students have worked with determination to give our best solution.
The journey was filled with long coding sessions, neverending group meetings and many very creative ways to approach this dataset.
We are proud to see where this project has lead us and are convinced you will appreciate the effort put into this presentation.
We hope going through this notebook will be a clear and smooth experience.
Enjoy the journey.
The datasets used can be downloaded here:
-
Network traffic: https://www.hs-coburg.de/forschung-kooperation/forschungsprojekte-oeffentlich/informationstechnologie/cidds-coburg-intrusion-detection-data-sets.html
-
Android Malware Detection: https://figshare.com/articles/Android_malware_dataset_for_machine_learning_2/5854653
PIPELINE:
NETWORK TRAFFIC
VISUALIZATION
- Train Set
- Test Set
CLASSIFIERS
-
Data Preparation and Manipulation
-
Data Preprocessing
-
Classification with Numerical Columns a) Prediction of 'Class' b) Prediction of 'Binary Class': Converting class attribute into binary c) Prediction of 'Attack type' c.i) Random Downsampling
-
Classification Introducing Additional Features a) Day of the Week b) Flags c) IP Address d) All 3 of them e) Counting total requests in a given amount of seconds
-
Clustering a) Data Preprocessing b) Kmeans c) DBScan
ANDROID MALWARE DETECTION