A Malware Propagation Prediction Method

Abstract

The invention belongs to the technical field of Internet applications, and in particular relates to a method for predicting the propagation of malicious software; the method includes acquiring user nodes and their interaction data in a database, and extracting the propagation attributes of the user nodes; using Doc2vec algorithm to propagate content from the user nodes The user behavior feature vector of the user node is learned from the composed paragraphs; the vectorization algorithm Tensor2vec based on tensor decomposition is used to learn the user node network structure feature vector from the malware propagation network; the malware is carried out in the graph convolutional neural network. Propagation prediction, and predict whether the malware is propagated to user nodes and the propagation trend of the malware; the present invention takes into account the problem of inaccurate calculation accuracy caused by data sparseness, and uses a tensor decomposition method to calculate the communication between user nodes. Infection intensity, and use the representation learning method to mine the spatial feature information of malware spread, which can effectively predict the spread of malware.

Description

technical field [0001] The invention belongs to the technical field of Internet applications, relates to network and information security technologies, and in particular relates to a method for predicting the spread of malicious software. Background technique [0002] The amount and harm of malware has increased dramatically in recent years, and its threat to the network and user nodes is considered one of the most significant risks in the coming years. Early malware was mainly limited to computer viruses, but with the development of the Internet and the diversification of network attacks, the concept of malware has transcended the traditional narrow concept, especially with the development of Advanced Persistent Threat (APT). ), Supply Chain Attacks (SCA), botnets and ransomware and other malware have emerged, and malware has highlighted its proprietary, controlled and destructive targets. [0003] Additionally, ransomware infection rates declined for the first time three ...

AI Technical Summary

Problems solved by technology

[0005] Based on the classic SIR propagation dynamics research malware propagation prediction is based on the network structure and user node attributes to extract features, but does not fully consider the impact of potential interactions between user nodes on the accuracy of feature extraction

Models based on machine learning or neural networks often ignore the diversity of user relationships and user behaviors in the malware propagation network, resulting in the need to explicitly extract the infection intensity between users and the influence of different malware

Images