Malware Prediction Using Deep Neural Network

Abstract

This research focuses on Malware Prediction techniques using different machine learning algorithms on Microsoft Telemetry Dataset. The malware industry continues to be a well-organized, well-funded market dedicated to evading traditional security measures. Once a computer is infected by malware, criminals can hurt consumers and enterprises in many ways. This project focuses on developing techniques to predict if a machine will soon be hit with malware, using machine learning. Large- and small-scale IT companies are under constant threat of malware and large sums of money is spent to protect these systems, in addition to regular security audits of the systems, because once the systems are infected by malware, there is a serious threat to the businesses and their customers in many ways including data theft. For these reasons, this project is proposed with the goal of being able to predict any malware activity using machine learning techniques, to thwart the threats before they could harm the systems. We have trained models using three different Machine Learning algorithms to compare the most accurate and robust algorithm for malware prediction. This kind of research is also relevant to our national needs as now a days, more and more reliance is being put on IT systems all over the world including Pakistan, because they are efficient and reliable. These systems range from simple bus ticket management system to National data registration system and border management systems, which hold very critical and sensitive national information. The security and integrity of these systems are a paramount concern for the authorities and if a successful algorithm and system is developed which could effectively analyze and report any threat to these systems before they happen, it will be of immense importance and utility. With our research we will be able to deliver such a system that would be able to predict malicious activities including malware attacks and viruses in IT systems.