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

A design method of multi-layer clustering fusion mechanism for multi-dimensional attribute data

A design method and multi-dimensional attribute technology, applied in computing, computer components, instruments, etc.

Active Publication Date: 2019-01-01
NANJING UNIV OF POSTS & TELECOMM
View PDF1 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a design method of a multi-layer clustering fusion mechanism for multi-dimensional attribute data. Data clustering fusion problem

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
  • A design method of multi-layer clustering fusion mechanism for multi-dimensional attribute data
  • A design method of multi-layer clustering fusion mechanism for multi-dimensional attribute data
  • A design method of multi-layer clustering fusion mechanism for multi-dimensional attribute data

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0085] 1. Architecture

[0086] Such as figure 1 As shown, the present invention analyzes the characteristics of each stage of multi-sensor information clustering and fusion, and first sets a threshold to filter and delete abnormal points in the analysis domain data. The information clustering fusion architecture based on multi-dimensional mixed attributes mainly includes four parts: extraction of optimal reference standards based on index attributes, gray relational cluster analysis, application of rough set theory, and probability statistics data level.

[0087] 2. Method flow

[0088]In the wireless sensor network, the invention uses a group of sensor nodes to collect different types of information at the same time for a certain target, and processes the data information to extract valuable knowledge. Since the data sensed by sensor nodes may be missing or uncertain, the collected data is first preprocessed and converted into a matrix format, and then thresholds are set ...

Embodiment 2

[0157] 1. Analyze domain data for gray relational clustering

[0158] The analysis domain data clustering processing planning process is as follows:

[0159] 1) Collect a group of sensor system nodes in a monitoring area to monitor targets, obtain sensing data, and convert it into a matrix format through preprocessing. X={X i |X i =(X i1 ,...,X im ,class)}i∈N is the comparison object set of analysis domain data, Y={Y i |Y i =(Y i1 ,...,Y im )}(i=1,2,...,p) is a known reference sequence set. By setting the threshold, filter and delete the abnormal data points of the analysis domain data.

[0160] 2) According to the characteristics of data index attributes, extract the optimal reference standard X 0 .

[0161] 3) According to the attributes of each characteristic index, normalize the data in the analysis domain, eliminate the impact of dimension (unit), and compress each data object in the analysis domain to the [0, 1] interval.

[0162] 4) Take the resolution coeff...

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 design method of a multi-layer clustering fusion mechanism for multi-dimensional attribute data. The method includes: step 1: converting the data set into a matrix form, and preprocessing the data; step 2: according to the attribute characteristics of the data index Extract the optimal reference standard, and normalize the data; Step 3: Calculate the gray correlation degree, generate a gray correlation degree similarity matrix, and then perform gray correlation degree clustering to obtain the initial clustering result; Step 4: According to the above steps 3 For the initial clustering results, use rough set theory to establish a decision table system; Step 5: Calculate the attribute importance information entropy value of each cluster member in the decision system; Step 6: Set weights for each cluster member ; Step 7: According to the calculated weight, use the probability method to calculate the probability of each data object in each class level, and select the class level where the maximum probability is obtained, which is the class level to which the data object belongs, and obtain the final Cluster fusion results.

Description

technical field [0001] The invention relates to a design method of a multi-layer cluster fusion mechanism oriented to multi-dimensional attribute data, and belongs to the technical field of data mining. Background technique [0002] Clustering fusion technology is applied to analyze and process, mine data, and extract useful knowledge for the irregularity and dispersion of data. The clustering fusion algorithm is an unsupervised machine learning algorithm. Unlike the supervised learning algorithm, it does not require prior knowledge of the distribution of the data set. The purpose of the clustering algorithm is to divide the data into several categories to reveal the real situation of the data distribution. Usually, the data collected by a group of sensor nodes generally has multiple mixed attributes, and the data of multiple attributes are clustered and fused to avoid the blind area of ​​single attribute data processing, improve the quality of multi-source information proc...

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 Patents(China)
IPC IPC(8): G06K9/62
CPCG06F18/232
Inventor 叶宁张迎亚黄海平沙超王汝传
Owner NANJING UNIV OF POSTS & TELECOMM
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