Deep Learning-based Malware Classification Methodology of Comprehensive Study

The increasing connectivity of devices and systems in today’s world has led to a significant increase in the spread of malware. The rapid advancements in technology have made it easier for cybercriminals to gain access to crucial information and disrupt mechanisms on a very large scale. With the widespread use of the internet, the proliferation of mobile devices, and the growing adoption of these devices, the opportunities for malware to spread have grown exponentially. Malware has the ability to recreate itself and spread across systems through various mechanisms. Malware can spread from any of the following: phishing emails, websites, infected USBs, remote access, and other removable media. As computers have become more integrated into nearly every aspect of our lives, it inadvertently leads to the evolution of malware that can spread more easily. We need a mechanism to differentiate the different kinds of malware by classifying them, so as to pre-emptively fight against the malware. This study uses a combination of convolutional neural network and spatial attention, a deep learning model that classifies 25 malware families using image-based methodologies.