Method for generating Fourier coefficient-based symbolic category set of multivariate time sequence

A multivariate time series and Fourier coefficient technology, applied in the field of time series data mining, can solve problems such as difficult to describe the trend of data changes in a segment, and achieve the avoidance of missing check behavior, smoothing of time domain time series, and removal of short-term noise effect

Inactive Publication Date: 2018-09-28
CHONGQING UNIV
View PDF4 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Disadvantages: The current symbolic time series data representation is only extracted from the mean value of the sequence segment, and there are feature extraction problems such as difficulty in describing the trend of data changes within the segment

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
  • Method for generating Fourier coefficient-based symbolic category set of multivariate time sequence
  • Method for generating Fourier coefficient-based symbolic category set of multivariate time sequence
  • Method for generating Fourier coefficient-based symbolic category set of multivariate time sequence

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0035] Example: such as figure 1 Shown; A multivariate time series is based on the Fourier coefficient symbolized category set generation method, which includes: the method steps are as follows:

[0036] S1: Obtain multivariate time series data;

[0037] S2: Preprocess the multivariate time series data to obtain a standard sequence with a Gaussian distribution with a mean of 0 and a variance of 1;

[0038] S3: Segment the multivariate time series using the segment aggregation approximate representation algorithm, and obtain all segment information of each sequence;

[0039] S4: Discrete Fourier transform is performed on the segmented data of each sequence to obtain the trend feature in the sequence segment represented by the Fourier coefficient;

[0040] S5: Use symbol aggregation approximate representation method to symbolize the sequence segment of multivariate time series, and the symbol and Fourier coefficient corresponding to each sequence segment are the complete symbo...

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 method for generating a Fourier coefficient-based symbolic category set of a multivariate time sequence. The method comprises the steps of: acquiring multivariate time sequence data; pre-processing the multivariate time sequence data to obtain a standard sequence of which a mean value of gaussian distribution is 0 and a variance is 1; performing piecewise processing on the multivariate time sequence through a PAA (Piecewise Aggregate Approximation) algorithm to obtain piecewise information of each sequence; performing DFT (Discrete Fourier Transform) on piecewise dataof the each sequence to obtain a trend characteristic in a sequence piece represented by a Fourier coefficient; performing symbolic representation on the sequence pieces of the multivariate time sequence through an SAX (Symbolic Aggregate Approximation) method, and enabling a symbol and the Fourier coefficient corresponding to the each sequence piece to form a complete symbolic category set of the sequence piece. The method of the invention has the beneficial effects that: dimensionality reduction can be performed on high-dimensional mass multivariate time sequence data, and advantages of theSAX can be maintained; and dimensionality reduction is performed through a frequency-domain filtering method, thus invariance of an Euclidean distance can be maintained.

Description

technical field [0001] The invention relates to the technical field of time series data mining, in particular to a method for generating a multivariate time series symbolized category set based on Fourier coefficients. Background technique [0002] In view of the large amount of data, mass, complexity, and high-dimensionality of time series data, researches on classification, clustering, similarity query, anomaly detection, and pattern mining have been carried out. , Internet and other fields have a wide range of applications. But most of the research is on the analysis of univariate time series data, and the research on multivariate time series is still relatively small. In multivariate time series time series data, how to effectively extract the characteristics of multivariate time series data and analyze the information and knowledge contained in it after dimensionality reduction has important theoretical and practical significance for scientific research and practical 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
Patent Type & Authority Applications(China)
IPC IPC(8): G06F17/30G06F17/14
CPCG06F17/141
Inventor 张可柴毅李媛赵晓航游丹妮
Owner CHONGQING UNIV
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