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

Adaptive density clustering method, storage medium and system

A density clustering and self-adaptive technology, applied in the field of cluster analysis, can solve problems such as clusters are easily merged, affect the effect of DBSCAN algorithm, and the same cluster is easy to be divided, etc., and achieve the effect of guaranteeing the effect.

Pending Publication Date: 2022-03-01
CHONGQING COLLEGE OF ELECTRONICS ENG
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The performance of the DBSCAN algorithm is affected by two important parameters. When clustering different data sets, it needs to be manually set. If a fixed Eps and MinPts are used each time, it is not adaptable to data sets with different degrees of sparseness. As a result, the same cluster in a low-density area is easy to be divided, or different clusters in a high-density area are easy to be merged, and the manual setting is only based on the results of repeated tests or experience. If the set Eps and MinPts are not appropriate, it will also Seriously affect the effect of DBSCAN algorithm

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
  • Adaptive density clustering method, storage medium and system
  • Adaptive density clustering method, storage medium and system
  • Adaptive density clustering method, storage medium and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0072] The embodiment is basically as attached figure 1 Shown: a kind of self-adaptive density clustering method, this method is applied to image segmentation in the present embodiment, comprises the following contents:

[0073] sup k Calculation steps: Calculate the natural eigenvalue sup of the data set S k ;Specifically:

[0074] Input data set S, S contains several data objects: S={x 1 ,x 2 ,...,x n-1 ,x n};

[0075] For data object x i ,x i ∈S, if there is a data object x j ,x j ∈S,x i ≠x j the sup k Nearest neighbor path experiences x i , and sup k Satisfied that the most outlier data object in S has the nearest neighbor path, then the current sup k is the natural eigenvalue:

[0076]

[0077] where s.t.x ∈ NN k (y) represents a restriction on x and y: x and y are natural nearest neighbors belonging to each other; natural nearest neighbors: for data object x i ,x i ∈S, if there is a data object x j ,x j ∈S,x i ≠x j The nearest neighbor path of ...

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 relates to the technical field of clustering analysis, in particular to a self-adaptive density clustering method, a storage medium and a system. The method comprises the following steps: a supk calculation step: calculating a natural feature value supk of a data set S; a calculation step: according to the supk, calculating a natural feature set Eps: according to a data object in the supk, obtaining Eps of different density areas in the S; and a clustering step: setting MinPts and Eps according to the Eps of different density areas in the supk and S, and starting a DBSCAN algorithm to perform clustering. According to the scheme, the parameters MinPts and Eps can be adaptively set, and the influence of data set density distribution on the DBSCAN is overcome, so that the clustering effect is ensured.

Description

technical field [0001] The invention relates to the technical field of cluster analysis, in particular to an adaptive density clustering method, a storage medium and a system. Background technique [0002] With the rapid development of communication technology, the growth of information data has broken through the exponential level, resulting in data excess and information explosion. Traditional data information processing technology has been unable to extract valuable information from massive data information. Therefore, in order to meet people's needs for data information processing, data mining has emerged in the era of big data and has become an important technology for processing massive data information. Data mining is classified into classification, valuation, prediction, grouping by relevance or association rules, and clustering. Among them, clustering is a method to automatically find and establish grouping rules, and divide similar samples into a cluster by judgin...

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): G06F16/901G06F16/906G06F16/953
CPCG06F16/9027G06F16/906G06F16/953
Inventor 卢建云李腾路亚李士果绍俊明宁丹
Owner CHONGQING COLLEGE OF ELECTRONICS ENG
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