Malware Detection Software Powered by Machine Learning

Abstract

Malicious software is overflowing in a world of countless computer users, who are continuously faced with these threats from various sources like the internet, local networks, and portable drives. The continued evolution of Malware is potentially a major threat in this cyber world. The current paper aims to create software powered by machine learning to detect whether a given software is malicious or not, before the installation of the software in the system. This task will be accomplished by utilizing machine learning algorithm called Random forest classifier, which is a type of supervised learning algorithm and will try to detect malware without relying on any Signature-based traditional techniques which are processor-intensive and efficient due to large amount of malware being made on day to day basics rather rely on Static analysis using PE file format with the help of feature extraction and build an effective, processor efficient malware detection software with high accuracy and low false-positive rate.