Network data intrusion detection method and system, terminal and storage medium

Abstract

The invention provides a network data intrusion detection method and system, a terminal and a storage medium, and the method comprises the steps: collecting network transmission data as sample data, dividing the sample data into a plurality of groups, wherein the network transmission data comprise normal data and intrusion data; removing redundant features, irrelevant to the preset target category, in each group through an FCBF algorithm, and the data groups with the redundant features removed serving as parental line sets; carrying out single-point genetic cross processing on the parental line set, and screening out an initial set with the highest relevancy with the target category from the parental line set and the genetic sub-line set; calculating a variation set of the initial set by utilizing a genetic variation algorithm, and iteratively screening out a data set with the highest relevancy with the target category by comparing the relevancy between the initial set and the target category and the relevancy between the variation set and the target category; and taking the data set with the highest relevancy with the target category as a feature subset, and performing feature extraction and feature detection on the network transmission data according to the feature subset.

Description

technical field [0001] The invention relates to the technical field of network intrusion detection, in particular to a network data intrusion detection method, system, terminal and storage medium. Background technique [0002] With the rapid development of big data and cloud computing, network intrusion methods are characterized by concealment and silence, and people's demand for network security is increasing. However, with the increase of data volume, the efficiency of mainstream network intrusion models for intrusion identification is getting lower and lower. Because these data are not only large in scale, but also high in dimension, there is a lot of redundant information, and these irrelevant information will greatly reduce the effect of intrusion detection. Feature selection is an effective means to solve this problem. A good feature selection algorithm can effectively eliminate redundant features or noise data in the classification data, and improve the speed and acc...

AI Technical Summary

Problems solved by technology

However, the feature subset obtained in this way is prone to overfitting. Once the sample set changes, the feature subset will not be applicable to the new sample set. Using the original feature subset for classification will lead to a decrease in classification accuracy.

Images