Malware-Analysis-with-Machine-Learning

Project aims to predict if a software is malware or not by using system call sequences in different window sizes.

Basic Usage

.m files should be opened with MATLAB.
Datasets are seperated by window sizes, data_5 has window size as 5 etc.
Last columns in datasets are 1 for the malicious software, 0 for not.
Datasets should be in the same path with the code, if not path of the datasets must be set manually or fixed in the code.

data = readtable('PATH OF THE DATASET');

For plotting the result in bar plot, algorithm names must be uniqe for each row.

Accuracy results of different algorithms and customizations are stored in Results folder in results.csv.