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

Clustering center determining method, determining system and clustering method

A technique for determining the method and clustering center, applied to instruments, character and pattern recognition, computer components, etc., to overcome time-consuming clustering and inaccurate clustering results

Inactive Publication Date: 2017-12-15
HUBEI UNIV OF TECH +1
View PDF0 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] However, the initial cluster center of the traditional K-means clustering algorithm is randomly selected

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
  • Clustering center determining method, determining system and clustering method
  • Clustering center determining method, determining system and clustering method
  • Clustering center determining method, determining system and clustering method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 2

[0116] Such as image 3 As shown, a system for determining cluster centers includes:

[0117] The water wave group construction module 201 is configured to construct a water wave group including a plurality of water waves, and randomly initialize the position, wave height, and wavelength of each water wave, wherein each water wave includes m cluster centers;

[0118] The propagation module 202 is configured to perform propagation processing on each of the water waves in the water wave group;

[0119] The first judgment module 203 is configured to respectively judge whether the fitness value of each water wave after the propagation processing is greater than the fitness value of the water wave before the propagation processing, and obtain a first judgment result;

[0120] The replacement processing module 204 is configured to replace the propagation processing in the water wave group with the water wave after the propagation processing if the first determination result indicates that th...

Embodiment 3

[0126] Such as Figure 4 As shown, a clustering method includes:

[0127] Step 301: Obtain a data set to be clustered and an optimal cluster center. The optimal cluster center is the optimal cluster center determined according to the determination method in Embodiment 1, wherein the data set to be clustered contains n data sets, the number of clusters is k;

[0128] Step 302: Perform cluster division on each data in the data set according to the closest distance criterion;

[0129] Step 303: Determine whether the termination condition is met;

[0130] If yes, go to step 304;

[0131] Otherwise, return to step 302;

[0132] Step 304: Output the optimal clustering result.

[0133] In this embodiment, the termination condition in step 303 may be set as: the current iteration number reaches the set maximum iteration number.

[0134] In the K-means clustering process, the closest distance criterion criterion is used to divide samples, that is, when the cluster center is determined, each sample...

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 clustering center determining method and a clustering center determining system. The method includes the following steps that: a water wave group including a plurality of water waves is constructed; propagation processing is performed on each water wave in the water wave group; and whether the fitness value of each water wave obtained after the propagation processing is performed is greater than the fitness value of the corresponding water wave obtained before the propagation processing is performed judged; if the fitness value of each water wave obtained after the propagation processing is performed is greater than the fitness value of the corresponding water wave obtained before the propagation processing is performed, the water waves obtained after the propagation processing are adopted to replace the corresponding water waves which are obtained before the propagation processing; if the fitness value of each water wave obtained after the propagation processing is performed is not greater than the fitness value of the corresponding water wave obtained before the propagation processing, the wave height and wave length of the water waves in the water wave group which are obtained before the propagation processing are updated, and the locations of the water waves which are obtained before the propagation processing are kept unchanged; the wave group and the number of iterations are updated; whether the current number of iterations is smaller than a set evolution algebra is judged; if the current number of iterations is smaller than the set evolution algebra, the method shifts to previous steps, propagation processing is performed on the water waves in the updated water wave group; and if the current number of iterations is not smaller than the set evolution algebra, water waves in the updated water wave group, which have the largest fitness value, are selected as optimal water waves. According to the method and system provided by the invention, from the perspective of optimization search, a clustering center is optimized from generation to generation by means of propagation processing, and the clustering center can be approximate to an optimal clustering center.

Description

Technical field [0001] The present invention relates to the field of big data mining, in particular to a clustering method and system. Background technique [0002] With the development of big data technology, the amount of data generated has increased sharply. Traditional data processing methods can no longer meet the requirements. Cluster analysis as a big data mining technology has once again become a research hotspot. Cluster analysis is an important unsupervised learning method. The purpose of cluster analysis is to find the structure hidden in the data, and according to a certain similarity measure, as far as possible to put the data with the same nature into the same category. [0003] However, the initial cluster centers of the traditional K-means clustering algorithm are randomly selected. For a clustering algorithm with multiple iterations, the cluster center of this iteration is obtained by updating the center of the previous iteration. Therefore, as the calculation bas...

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): G06K9/62
CPCG06F18/23213
Inventor 王春枝刘川叶志伟胡继雄陈宏伟刘伟宗欣露苏军严灵毓陈颖哲任紫扉王俊
Owner HUBEI UNIV OF TECH
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