This repository contains the source code of a project that aims to analyze the results of Java implementations of Naive-Bayes and SVM classifiers over the Drebin Android malware dataset.
The main file (Homework1.java) allows to create in a random way (with or without percentage control of malware's incidence) a fixed number of sample subsets of the main dataset that will be trained and tested with a Naive-Bayes or SVM classifier. All the procedures (including a debugging one based on an Evaluator class that can report the features of a sample) can be controlled by modifying the private fields of the Homework1 class.
Results of each test are automatically printed on resulting txt log files.