Unlock instant, AI-driven research and patent intelligence for your innovation.

Network attack detection system and method based on two-stage learning model

A technology of network attack and learning model, applied in the field of two-stage detection system of network attack data preprocessing and network attack data identification, can solve the problems of long model training time, increase the difficulty of network attack detection model and system deployment, etc. volume effect

Active Publication Date: 2021-12-03
HANGZHOU DIANZI UNIV
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

At the same time, the existing deep learning models for network attack detection often believe that the learning model is powerful, and feature preprocessing is not performed, which leads to a long training time for the model and increases the difficulty of deploying the network attack detection model and system

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Network attack detection system and method based on two-stage learning model
  • Network attack detection system and method based on two-stage learning model
  • Network attack detection system and method based on two-stage learning model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0041] like figure 1 As shown, the network attack detection method based on the two-stage learning model in this embodiment is carried out in the following steps:

[0042] Stage 1: Reduction of the feature dimension of the network dataset; in this stage, the feature subset of the network dataset is used as an inseparable unit for feature combination evaluation, and the feature dimension reduction of the network dataset is realized; the details are as follows:

[0043] Step 1.1, preprocessing of massive network data. Firstly, the data instances that exceed the limit of the missing feature threshold are screened out; then, the features with low information content are deleted in the network data set after the initial screening, such as serial number, timestamp, etc.; finally, the non-numeric data in the network data set The features are one-hot encoded and mapped to binary vectors;

[0044] Step 1.2, construct a feature subset evaluation function for the network dataset. Base...

Embodiment 2

[0053] like Figure 4 As shown, the network attack detection system based on the two-stage learning model in this embodiment includes the following modules:

[0054] The feature dimension reduction module of the network dataset: this module evaluates the feature combination of the feature subset of the network dataset as an inseparable unit, and realizes the feature dimension reduction of the network dataset; specifically, it includes the following sub-modules:

[0055] Preprocessing module of massive network data: preprocessing of massive network data. Firstly, the data instances that exceed the limit of the missing feature threshold are screened out; then, the features with low information content are deleted in the network data set after the initial screening, such as serial number, timestamp, etc.; finally, the non-numeric data in the network data set The features are one-hot encoded and mapped to binary vectors;

[0056] Feature Subset Evaluation Function Building Block...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a network attack detection system and method based on a two-stage learning model, and the method comprises the following steps: 1, enabling a feature subset of a network data set to serve as an inseparable unit, carrying out the evaluation of feature combination, and achieving the feature dimension reduction of the network data set; and 2, taking the reduced network data set as training data, and realizing a network attack real-time detection model by using a deep learning technology. According to the two-stage network attack detection technical scheme, the feature combination effect of network high-risk data is fully considered, and the feature selection technology, the evolutionary search technology and the deep learning model are combined for the accuracy and timeliness which need to be guaranteed by network attack detection, thus to improve the recognition precision of network attack detection, and greatly shorten the model training time.

Description

technical field [0001] The invention belongs to the technical field of network attack detection, in particular to a two-stage detection system and method for network attack data preprocessing and network attack data identification. Background technique [0002] How to effectively implement network attack detection in the context of the Internet of Everything is one of the key issues faced in the big data environment. In recent years, various recognition methods based on deep neural networks have been widely used in network attack detection systems. Compared with traditional machine learning methods, deep neural network models can often achieve higher detection accuracy. The most critical factor affecting various network attack detection models is the use of data features of high-dimensional network attacks. Therefore, the first basic task of extracting valuable information from large-scale high-dimensional network attack data is to find the key features of network attack dat...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): H04L29/06G06N3/08G06N3/04G06K9/62
CPCH04L63/1416G06N3/08G06N3/045G06F18/214
Inventor 滕旭阳张云啸何美霖毕美华仇兆炀
Owner HANGZHOU DIANZI UNIV