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

High-precision time sequence prediction method for inhibiting data leakage

A technology of time series and data leakage, which is applied in the field of communication technology and computer, can solve the problems of reducing the applicability of prediction models, data leakage, and predictions that do not conform to real applications, etc., and achieve the effect of suppressing data leakage and improving prediction performance

Pending Publication Date: 2022-08-05
SHENYANG LIGONG UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, most of the current methods use smoothing or noise reduction of the overall time series, and then divide the training set and test set, which leads to the problem of data leakage, makes the prediction not conform to the real application, and reduces the applicability of the prediction model

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
  • High-precision time sequence prediction method for inhibiting data leakage
  • High-precision time sequence prediction method for inhibiting data leakage
  • High-precision time sequence prediction method for inhibiting data leakage

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0012] Step 1: Collect time series. Suppose the sequence x(n) of length N is polluted by noise u(n), n=1,2,...,N, the collected noisy sequence can be expressed as:

[0013] y(n)=x(n)+u(n) (1).

[0014] Step 2: Perform VMD processing on the time series y(n). VMD can decompose the input sequence y(n) into a different number of subsequences with limited bandwidth. The subsequences are Intrinsic Mode Function (IMF) components, and these IMF components can reproduce the original input according to their sparsity. sequence, as shown in Equation (2) and Equation (3), where Y w (n) is the decomposed IMF component, F[ ] is the VMD decomposition function, and A is the parameter matrix including the decomposition scale K, penalty factor α, noise tolerance τ and discrimination accuracy ε.

[0015] Y w (n)=F[y(n)] A ,w=1,2,...,K(2).

[0016]

[0017] It has been proved by experiments that the values ​​of parameters τ and ε have little influence on the decomposition results, usuall...

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 high-precision time sequence prediction method for inhibiting data leakage, belongs to the technical field of communication technologies and computers, takes a noise reduction correlation theory and a deep learning prediction model modeling thought as a theoretical basis, considers the characteristics of high noise, non-stability, nonlinearity and the like of a time sequence, and provides a high-precision time sequence prediction method for reducing the influence of noise on time sequence prediction. The method comprises the following steps of: performing multiple variational mode decomposition (VMD) processing on a time sequence through overlapping slices, improving a noise reduction threshold function, and performing noise reduction processing on the decomposed time sequence. Furthermore, a deep learning modeling thought is introduced, and a neural network multi-step prediction model is established, so that the prediction performance of the time sequence is improved. The method is suitable for a time sequence multi-step prediction system and device with noise characteristics.

Description

technical field [0001] The present invention belongs to the fields of communication technology and computer technology, in particular to the high-precision time series prediction technology, which can suppress data leakage and realize multi-step prediction. Background technique [0002] Many applications in production and life involve time series, such as stock price index, traffic flow, air pollution index, rainfall, EEG signals and sunspots. The use of time series prediction can reveal the internal evolution law of complex systems. Therefore, time series prediction is an important means to explore the mechanism of complex systems and establish system models. However, because time series usually have the characteristics of high noise and nonlinearity, it increases the difficulty of time series forecasting. There are bound to be influencing factors (referred to as noise) in the time series, which will not only reduce the quality of the time series, but also lose effective i...

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): G06F21/64G06K9/62G06N3/04G06N3/08
CPCG06F21/64G06N3/08G06N3/044G06F18/214Y02A90/10
Inventor 刘芳陈立志冯永新
Owner SHENYANG LIGONG UNIV
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