A network risk control method based on micro-state prediction

Abstract

The invention discloses a network risk control method based on microstate prediction. The method is characterized in comprising the steps of 1), finding a risk propagation source; 2), establishing a risk propagation micro model for a target network and obtaining a probability estimation value that each user node is in an infected state at a current moment under the micro model; 3), obtaining security states and risk propagation paths of user nodes in the target network through adoption of a sensor observation model and reestablishing a topological structure of the target network in which risks exist through combination of strategies, thereby obtaining the probability value that each user node is in the infected state at the current moment; and 4), comparing the probability values with a preset threshold value, and screening to obtain a control user node set for control. According to the method provided by the invention, the risk propagation can be effectively controlled, and the better risk control effect can be realized under the condition that fewer user nodes are immunized.

Description

technical field [0001] The invention relates to the field of network information security, in particular to a network risk control method based on micro-state prediction. Background technique [0002] The widespread use of the Internet has made us more vulnerable to a variety of cyber risks. For example, rumors spread freely in social networks, computer viruses in the Internet infect a large number of network hosts, and network isolation failures in smart grids can cause large-scale power outages. Every year, untold amounts of financial and social wealth are lost due to these cyber risks. For example, rumors spread widely on social networks since 2013 that "Shuang Wai Wai, Wahaha AD calcium milk, etc. all contain botulinum toxin, and are now urgently recalled." have caused a total of more than 2 billion economic losses to Wahaha Group. [0003] Internet rumors, computer viruses, and smart grid failures are all network risks that spread in different networks. In order to de...

AI Technical Summary

Problems solved by technology

Second, the key nodes in the static topological structure are not necessarily the important topological nodes that can affect the propagation at the current moment. For example, in the two special scenarios mentioned above, it is meaningless to control these so-called key topological nodes

In other words, the original method is too idealistic in the design of network risk control, and does not take into account the real-time dynamic changes of risk diffusion

Images