Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Two-weighted online extreme learning machine-based network intrusion detection method

A network intrusion detection and extreme learning machine technology, applied in electrical components, transmission systems, etc., can solve problems such as class imbalance, and achieve the effect of solving class imbalance, ensuring classification accuracy and robustness, and improving classification accuracy.

Inactive Publication Date: 2016-11-23
LIAONING NORMAL UNIVERSITY
View PDF1 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The present invention aims to solve the above-mentioned technical problems existing in the prior art, and provides a network intrusion detection method based on a double-weighted online extreme learning machine that can simultaneously deal with the problems of class imbalance and concept drift in data and effectively improve detection accuracy

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
  • Two-weighted online extreme learning machine-based network intrusion detection method
  • Two-weighted online extreme learning machine-based network intrusion detection method
  • Two-weighted online extreme learning machine-based network intrusion detection method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0024] A network intrusion detection method based on a double-weighted online extreme learning machine, characterized in that:

[0025] a. The online extreme learning machine is carried out according to the following steps:

[0026] Step 1: Initialize

[0027] 1.1 From the training set D randomly selected from n 0 samples as the initial training set D 0， , the present invention selects the training set D 5% of the data is used as the initial training set, and the remaining data is divided into blocks, and different block sizes are used for different data. In order to ensure that the imbalance rate of the test set is the same as that of the entire data set, according to the size of the imbalance rate, the present invention selects 20% of the remaining 95% data as test data, and 80% of the data as training data.

[0028] 1.2 Randomly assign input weights and thresholds;

[0029] 1.3 Utilization For the initial training sample set D 0 Compute the initial intermed...

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 two-weighted online extreme learning machine-based network intrusion detection method. The method comprises the following steps: in an initial training stage, randomly selecting samples from a dataset as initial training samples, randomly allocating weights and threshold values, further calculating a weight on a time level, training an initial probabilistic neural network, obtaining scores of each sample in each type, calculating degrees of membership to obtain a weight on a space level, and finally combining the weights on the time level and the space level to obtain a final initial weight; in a continuous learning stage, for each block of new arriving data, updating and diagonalizing the weight on the time level, updating the probabilistic neural network to obtain a score of the current block, further updating the weight on the space level, and finally combining the updated weights on the time level and the space level to obtain a final output weight.

Description

technical field [0001] The invention relates to the field of data mining, in particular to a network intrusion detection method based on a double-weighted online extreme learning machine that can simultaneously deal with the problems of class imbalance and concept drift in data and effectively improve detection accuracy. Background technique [0002] With the rapid development of computer networks, network communication has penetrated into all walks of life, played a key role in the development of human society, and influenced and changed people's lives. Although the network brings convenience to people, it also brings various security problems because of network intrusion. For example: network hackers can break through confidential documents, steal bank deposits, tamper with and destroy data blocks, and so on. At present, for a large amount of data generated by network intrusion, useful knowledge is basically extracted from it with the help of relevant methods of data mini...

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
Patent Type & Authority Applications(China)
IPC IPC(8): H04L29/06
CPCH04L63/14H04L63/1416H04L63/1433
Inventor 张永刘文哲刘博
Owner LIAONING NORMAL UNIVERSITY
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com