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

Industrial internet intrusion detection data set processing method based on D-N

An industrial Internet and intrusion detection technology, applied in the field of KDD99 data set processing, can solve problems such as missing data sets

Pending Publication Date: 2022-01-14
JILIN UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The present invention is proposed in order to solve the high requirements of the integrated learning algorithm on the data structure of the data set and the lack of industrial Internet intrusion detection data set. The D-N-based industrial Internet intrusion detection data set processing algorithm through data cleaning, The three steps of data normalization realize the data analysis and organization of the data set

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
  • Industrial internet intrusion detection data set processing method based on D-N
  • Industrial internet intrusion detection data set processing method based on D-N
  • Industrial internet intrusion detection data set processing method based on D-N

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0010] (1) Input the data set D that needs to be processed, and traverse all the data labels l of the data set D 1 , l 2 ,... l n .

[0011] (2) According to the traversal results of the data labels of the data set D, an empty table E with the header and the data label sequence and name of the data set D is completely consistent, that is, the data cleaning pool.

[0012] (3) Each data label l in the data cleaning pool E 1 , l 2 ,... l n Enter the value v of the data label to be processed respectively 11 , v 12 ,...,v 1m ; v 21 , v 22 ,...,v 2m ;...;v n1 , v n2 ,...,v nm , and the processing method M of each item, the update data cleaning pool is E f .

[0013] (4) Traverse the data set D in the order of row by row and then column by column, and compare the data cleaning pool E f , process the data tags that need to be processed in the processing mode M, and obtain the data set D after traversal processing f .

[0014] (5) Traverse the data set D in the order ...

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 an industrial internet intrusion detection data set processing method based on D-N. The algorithm improves the problems that, when the existing ensemble learning algorithm solves the industrial internet intrusion detection problem, redundant data items in a data set cause poor generalization performance of a trained ensemble learning model, and some types of data labels in the data set cannot be identified by an individual learner of ensemble learning, and some types of data labels in a data set are wrongly recognized by an individual learner of integrated learning, so that a trained integrated learning model is low in detection precision. A new method is provided for processing a training data set and a verification data set when an integrated learning algorithm is used for solving the industrial internet intrusion detection problem.

Description

technical field [0001] The invention relates to the processing of data sets, data cleaning, discrete-normalization mathematical method (D-N algorithm), classification of integrated learning algorithms and their application fields, especially in the integrated learning algorithms used to realize industrial Internet intrusion detection based on The KDD99 dataset of the CART-AMV algorithm is being processed. Background technique [0002] The emergence of integrated learning algorithms has improved the complex and cumbersome single algorithm process in machine learning. By building a large number of individual learners with simple algorithms and various types, the algorithm complexity and cost of machine learning can be effectively reduced. This is integrated learning. Advantages of class algorithms. Its defect is that in the training of individual learners, it strongly depends on the training data set used. The quality of the data structure of the training data set directly a...

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): G06F16/215G06F16/2458G06F16/28G06N20/20
CPCG06F16/215G06F16/2462G06F16/285G06N20/20
Inventor 刘明山石伟诚周原韦晓宇
Owner JILIN UNIV