Malicious software classification method based on Markov graph and deep learning

Abstract

The invention discloses a malicious software classification method based on a Markov graph and deep learning, and the method comprises the steps: firstly carrying out the statistics of the byte transfer frequency of malicious software, converting the byte transfer frequency into a Markov image, and completing the classification of the malicious software through a deep convolution neural network, wherein the designed deep convolutional neural network structure is redesigned on the basis of VGG16, the depths of convolutional layers and pooling layers of the deep convolutional neural network structure are the same as those of the VGG16, and the deep convolutional neural network structure comprises 13 convolutional layers and 5 pooling layers in total. Different from the VGG16, the deep convolutional neural network designed by the invention only has one full connection layer, and the output dimension of the deep convolutional neural network is 1024. The method is wide in applicability, andredundancy of byte information can be effectively reduced; and the method does not depend on a pre-trained model, and has higher classification accuracy.

Description

technical field [0001] The invention relates to the technical field of malware classification and deep learning, in particular to a malware classification method based on Markov graph and deep learning. Background technique [0002] Malware is a computer program that infiltrates and damages a computer without the user's consent. According to its behavior characteristics, it can be divided into viruses, spyware, worms, Trojan horses, rootkits and so on. With the popularity of the Internet, malicious software has become a profit tool and political weapon for criminals. For example, the industrial worm "Stuxnet" (Stuxnet), which was first detected by Kaspersky Lab in June 2010, has infected more than 45,000 networks around the world and destroyed nearly 20% of Iran's atomic centrifuges. In April 2017, the global outbreak of ransomware software "WannaCry" (WannaCry) affected at least 150 countries, infected more than 230,000 computers, and caused losses of US$8 billion. In Ma...

AI Technical Summary

Problems solved by technology

[0006] Aiming at the problem that traditional machine learning algorithms are limited by feature engineering, it is difficult to quickly process a large number of malicious samples, and the accuracy of existing classification methods based on malware grayscale images and deep learning is not high, the present invention provides a method based on Markov Graphs and Deep Learning for Malware Classification with High Classification Accuracy

Images