Feature screening system and method for network traffic attack detection based on learning automaton

Abstract

The invention discloses a feature screening system and method for network traffic attack detection based on a learning automaton. The system comprises a data preprocessing module, a setting module, alearning automaton module, a random environment module, a feature screening module and an attack detection module. According to the invention, by interacting with the classifer and learning the evolution of the automaton is learned, redundant features are removed one by one, finally, the optimal features are screened out to form an optimal feature subset, the problems that the network traffic datasize is large, and the dimension is high are effectively solved, the network traffic attack detection efficiency can be effectively improved through the screened features, and the method can be applied to environments of large-scale networks such as a power grid industrial control network and the like.

Description

technical field [0001] The present invention relates to a technology in the field of feature screening, in particular to a feature screening system and method for network traffic attack detection based on learning automata. Background technique [0002] With the continuous change of attack methods, attack detection technology based on known attack characteristics can no longer protect the network from attacks, and it is very necessary to conduct attack detection on network traffic. The attack network traffic packet consists of massive traffic data, which records all activities and behaviors of power grid terminals. By analyzing and integrating these network traffic packets, features can be extracted from them to discover attacks. However, due to the huge amount of network traffic, in order to achieve attack identification, real-time processing must be achieved, which requires high efficiency of detection algorithms. Traditional neural network learning methods and most mach...

AI Technical Summary

Problems solved by technology

[0003] Aiming at the above-mentioned deficiencies in the prior art, the present invention proposes a feature screening system and method for network traffic attack detection based on learning automata, and innovatively uses the idea of ​​learning automata for feature screening, which not only effectively solves the problem of network traffic The problem of large data volume and high dimensionality also takes into account that each network traffic feature is discarded with equal probability. By gradually updating the probability, the removed features are ensured to be sufficiently redundant. By using these optimal features for attack detection, effective Improve the accuracy of network traffic attack detection, and have a good detection rate for network attacks

Images