Malicious code detection method based on graph convolutional neural network
The invention discloses a malicious code detection method based on a graph convolutional neural network, which comprises the following steps of: 1, obtaining network traffic data through a traffic capture program, and extracting an effective load of the network traffic; 2, defining an effective load...
Saved in:
Main Authors | , |
---|---|
Format | Patent |
Language | Chinese English |
Published |
07.11.2023
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | The invention discloses a malicious code detection method based on a graph convolutional neural network, which comprises the following steps of: 1, obtaining network traffic data through a traffic capture program, and extracting an effective load of the network traffic; 2, defining an effective load matrix: defining the effective load matrix of the network data packet as (aij) 255 * 255; wherein aij is equal to sum (i, j), and sum (i, j) is the adjacent occurrence frequency of the effective load i and the effective load j; 3, training a graph convolutional neural network model: inputting training data into a graph convolutional neural network, and iteratively spreading neighbor information by using a first-order local convolution operation; and 4, malicious code detection: inputting a to-be-detected code into the trained graph convolutional network model, and detecting the malicious code. By using the technology, the malicious codes can be quickly and accurately identified and analyzed, so that the security p |
---|---|
Bibliography: | Application Number: CN202310985147 |