Overview

Nowadays mobile devices are used everywhere in our daily life providing many valuable services, but on the other hand we are assisting to a rapid growth of malware designed for mobile devices. Due to the great threats and damage that malware can cause to users, Android malware detection has become increasingly important in the cybersecurity domain and lots of studies and researches have been carried out in order to find new methods and approaches to discover malicious patterns in mobile applications.
In this context, Machine Learning (ML) has became a very successful way to detect and classify malware. This is because, using standard ML classification algorithms is possible to automatically learn the features that distinguish malware families and automate the detection process.
This project describes an innovative approach to malware detection and family classification that has been carried out on the DREBIN dataset using the Naive Bayes machine learning model for text classification, implemented through the NaiveBayesText Python class.
For further information, see the documentation and the obtained results described in the ml-malware-analysis.pdf report, and the Python code in the project repository.