TFE_Malware_Visualization_and_Classification

This master thesis is interested in two steps :

1. The generation of new visualization methods for malwares and cleanwares, fast and memory-saver, using only raw sample data
2. The classification of malwares using the visualizations and MLP or CNN as classifier

For the experiments, different datasets have been used, their structure and their behavior on the results have been analyzed and we tried to add cleanwares to see the results.

Description of the files

• Plots/ : contains the files written to generate the plots of the master thesis
• Results/ : contains examples of results for visualizations, confusion matrix and family plots
• Utils/ : contains useful functions for this work
• MLP.py : implementation of our MLP
• binary2image.py : generator for visualizations of malwares
• classifier.py : implementation of our CNN on a classic dataset
• classifier_kaggle.py : implementation of our CNN for the public Kaggle dataset
• classifier_paper.py : implementation of our CNN to compare with the literature

Contact

It will be a pleasure to help you in case of problems or questions : [email protected]