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

Multi-relation network-based MNMF clustering method of multi-variable time sequences

A multi-relational network and time series technology, applied in the field of time series clustering in data mining, can solve problems such as poor multivariate time series clustering effect

Active Publication Date: 2018-11-16
YUNNAN UNIV
View PDF4 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] In order to achieve the above object, the present invention provides a method of MNMF clustering multivariate time series based on a multi-relational network, which solves the complex relationship between individual sequences in the multivariate time series of process monitoring and control in the prior art. The problem of poor clustering effect of multivariate time series

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
  • Multi-relation network-based MNMF clustering method of multi-variable time sequences
  • Multi-relation network-based MNMF clustering method of multi-variable time sequences
  • Multi-relation network-based MNMF clustering method of multi-variable time sequences

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0036] Table 1 gives the 3-variable time series data of temperature, air pressure and humidity in 12 regions, and each variable contains 8 observations. 3-variable time series {X for 12 regions 1 ,X 2 ,...,X 12} into three clusters, the first cluster contains {X 1 ,X 2 ,X 3 ,X 4}, the second cluster contains {X 5 ,X 6 ,X 7 ,X 8}, the third cluster contains {X 9 ,X 10 ,X 11 ,X 12}. in Table 1 Respectively represent the temperature, air pressure and humidity of i region.

[0037] Table 1. 3-variable time series data and their cluster labels for 12 regions

[0038]

[0039]

[0040] For a 3-variable time series {X for 12 regions 1 ,X 2 ,...,X 12} clustering, first normalize j=1,2,3 for j=1,2,3, and then calculate the Euclidean distance D between time series for different variables j ,j=1,2,3, the result is as follows:

[0041]

[0042]

[0043]

[0044]

[0045]

[0046]

[0047] Take ε 1 =1.4,ε 2 =0.8,ε 3 =1.6, then according ...

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 multi-relation network-based MNMF clustering method of multi-variable time sequences. The multi-variable time sequences are converted into a multi-relation network G; and themulti-relation network G is jointly decomposed through MNMF to obtain clustering results of the multi-variable time sequences. The method of the invention effectively fuses complex relations in and among variables into a clustering process, and improves clustering performance.

Description

technical field [0001] The invention belongs to the technical field of time series clustering in data mining, and relates to the conversion of multivariate time series to multi-relational networks and multiple non-negative matrix decomposition (MNMF) in multi-relational networks. Background technique [0002] With the rapid growth of digital information sources, large amounts of time-series data, such as personal health trajectories, climate data, socioeconomic indicators, etc., are being continuously generated and collected. Mining these data helps to discover hidden knowledge and information, such as temporal associations, community behavior patterns, etc. In recent years, the mining of time series data has attracted the attention of many researchers. [0003] Time series clustering is to divide the time series into several clusters, the sequences in the same cluster are highly similar, and the sequences in different clusters are low in similarity. This is a fundamental ...

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): G06K9/62G06F17/30
CPCG06F18/232
Inventor 周丽华杜国王赵丽红王丽珍陈红梅
Owner YUNNAN 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