Review of Malware Detection Using Deep Learning

Abstract

AbstractMalware is a program which is designed to harm or exploit any programmable device, service, or network by containing malicious code in it. Malware detection refers to the process of detecting the presence of malware on a system or whether a specific program is malicious or benign. Most common types of malware are rootkits, spyware, Trojan horses, viruses, and worms. Malware detection systems act as an early warning system for any organization or individual to prevent confidential information from getting disclosed. In this paper, deep neural network (DNN) is used to learn the features that optimally represent the given training data. In this, deep neural network learns the features from portable executable files. Thus, deep neural network-based systems are effective to detect existing and unknown malware with low false positive rates. The proposed system detects whether a given portable executable (PE) file is malware or not using deep neural network (DNN).KeywordsMalwareDeep neural networkPortable executableNetwork securityAnti-malwareMalware detection systemAnti-malware system