L2 norm standardization and cosine theorem improvement-based elbow rule method

A cosine theorem and norm technology, applied in the field of elbow rule based on L2 norm normalization and cosine theorem improvement, can solve the problems of unfavorable promotion and use, personal subjectivity, etc., and achieve the goal of reducing the dependence of the optimal clustering number Effect

Inactive Publication Date: 2018-06-15
KUNMING UNIV OF SCI & TECH
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the method of the elbow rule often needs to present the number of clusters in the specified search range and the corresponding average degree of distortion in the form of a relationship curve, and then use the observation method to roughly estimate the elbow point on the relationship curve (best Cluster number), this method of identifying the optimal number of clusters by observation not only has personal subjectivity, but also is not conducive to its use in automation systems and its promotion and use in other fields

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
  • L2 norm standardization and cosine theorem improvement-based elbow rule method
  • L2 norm standardization and cosine theorem improvement-based elbow rule method
  • L2 norm standardization and cosine theorem improvement-based elbow rule method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0036] Embodiment 1: as Figure 1-5 As shown, the method based on the L2 norm normalization and the improved elbow rule of the law of cosines includes the following steps:

[0037] (1) Set the range to search for the best K value in the K-means clustering algorithm [K min , K max ];

[0038] (2) Use the K-means clustering algorithm to calculate the search range [K min , K max K in ] max -K min The average degree of distortion corresponding to the number of +1 clusters;

[0039] (3) For the calculated K max -K min +1 average distortion degree for L2 norm normalization;

[0040] (4) K after normalizing the L2 norm max -K min +1 Average Distortion Level vs. Search Range [K min , K max K in ] max -K min +1 clustering number packed into K max -K min +1 data point;

[0041] (5) Utilize the law of cosines to find the K packaged above max -K min The angle between every three adjacent data points in +1 data point;

[0042] (6) Find the obtained K max -K min - the...

example 1

[0059] Example 1: The specific steps of the method for automatically identifying the optimal K value in the elbow rule are as follows:

[0060] Step1. Set the range of the best K value to search for in the K-means clustering algorithm Range: [K min , K min +1,...,K max -1,K max ];specific:

[0061] Assuming that the K-means clustering algorithm is to search for K in the best K value range Range min = 1,K max =9, that is, the range of searching for the best K value is: [1, 2, 3, 4, 5, 6, 7, 8, 9]; the actual number of clusters contained in the sample data set involved in this example is 3 , that is, the actual number of clusters is in the search range Range of the best K value;

[0062] Step2. Initialize the number of clusters k=K min , and generate a length K max -K min +1 and the average distortion degree list MDL with all elements being 0; specifically:

[0063] Since K is assumed in Step1 min =1, so initialize k to 1, that is, k=1; since K is assumed in Step1 mi...

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 an L2 norm standardization and cosine theorem improvement-based elbow rule method, and belongs to the technical field of clustering analysis in data mining. The method comprises the steps of setting a range [Kmin-Kmax] of searching an optimal cluster number (an optimal K value) in a K-mean clustering algorithm; calculating average distortion degrees corresponding to Kmax-Kmin+1 clusters in the search range; and performing L2 norm standardization processing on the obtained Kmax-Kmin+1 average distortion degrees. Based on the K-mean clustering algorithm and elbow rules,the average distortion degrees obtained by the elbow rules are subjected to the L2 norm standardization processing and are further calculated by utilizing a cosine theorem, so that the optimal K valuein the search range is obtained; and the method can enable the obtained optimal K value to be more objective.

Description

technical field [0001] The method of L2 norm normalization and cosine theorem improved elbow rule, especially a combination of K-means clustering algorithm commonly used in cluster analysis in data mining, elbow rule and K-means clustering algorithm through L2 norm normalization The obtained average degree of distortion is normalized, the average degree of distortion processed after normalization and the number of clusters in the corresponding search range are packaged into data points, and then the law of cosines is used to find out every The method of identifying the best K value within the set search range by the angle between three adjacent data points, etc., belongs to the field of cluster analysis in data mining. Background technique [0002] With the rapid development of technologies such as information technology and the Internet of Things, the data generated by people has grown exponentially, and mankind has gradually entered the era of big data. At the same time, ...

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 付映雪石聪明王锋邓辉戴伟卫守林
Owner KUNMING UNIV OF SCI & TECH
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