Feature selection method for combined type unbalanced traffic classification

A feature selection method and traffic classification technology, applied in the fields of instruments, character and pattern recognition, digital transmission systems, etc., can solve the problems of easily ignoring the learning performance of small classes, not fully considering the distribution of data samples, etc., to reduce adverse effects. , the effect of reducing computational complexity and improving classification accuracy

Active Publication Date: 2019-07-05
CHONGQING UNIV OF POSTS & TELECOMM
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Most of the existing traditional feature selection methods aim to improve the classification accuracy without fully considering the distribution of data samples, and generally pursue the learning effect of large classes, and tend to ignore the learning performance of small classes

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
  • Feature selection method for combined type unbalanced traffic classification
  • Feature selection method for combined type unbalanced traffic classification
  • Feature selection method for combined type unbalanced traffic classification

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0042] This embodiment provides a feature selection method for combined type unbalanced traffic classification, please refer to figure 1 shown, including:

[0043] S1: Perform statistics on the network traffic sample data to obtain statistical results. The sample data corresponding to each piece of network traffic includes category information of the category to which the network traffic belongs and values ​​of multiple attribute features.

[0044] Optionally, the traffic sample data is data in Moore's public dataset. Specifically, you can download the Moore public dataset from the Internet, randomly select 2 / 3 of all data streams as the training set, and the remaining 1 / 3 as the test set. Each application category and the number of data streams in the dataset are shown in Table 1 below:

[0045] Table 1

[0046]

[0047] The sample data corresponding to each piece of network traffic includes category information of the category to which the network traffic belongs and t...

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 feature selection method for combined type class unbalanced flow classification. calculating the relevancy between the category and the feature by using a non-search algorithm and a weighted symmetric uncertainty WSU; filtering redundant features according to the WSU between the features; a first target feature set is obtained; the calculation complexity of subsequent feature subset screening can be obviously reduced; and then further reducing the dimensionality of the features by adopting an SFS algorithm until the number of the features is increased to a specified dimensionality, so that adverse effects on network traffic classification caused by a sample distribution imbalance problem can be reduced, a feature set with strong distinguishing capability is selected, and the classification precision of the network traffic can be remarkably improved.

Description

technical field [0001] The present invention relates to the technical field of network traffic classification, and more specifically, relates to a feature selection method for combined type unbalanced traffic classification. Background technique [0002] With the rapid development of the Internet, the network coverage continues to expand, and the types of network applications continue to increase. While these changes bring convenience to people's lives, they also bring great challenges to network operation and management. Network researchers have proposed a series of measures to ensure the healthy operation of the network, but whether it is to realize efficient service carrying based on user needs, or to expand and transform the existing network according to the development trend of network applications, it is necessary to Various applications for accurate classification and identification. In addition, in fields such as intrusion detection, network traffic classification ...

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): H04L12/24G06K9/62
CPCH04L41/142G06F18/211G06F18/2415
Inventor 唐宏刘丹姚立霜王云锋裴作飞
Owner CHONGQING UNIV OF POSTS & TELECOMM
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap