Clustering method, system and medium for automatically confirming number of clusters based on coefficient of variation

Abstract

The invention discloses a clustering method, system and medium for automatically confirming the number of clusters based on the coefficient of variation, calculates the density value of each data point in the data set, calculates the density index according to the density value, and selects the data point with the largest density index as the first cluster. Class center; calculate the shortest distance between each data point and the current existing cluster center, and then calculate the probability of each data point being selected as the cluster center according to the shortest distance, and pre-select the cluster center according to the roulette method; until the selection Set a cluster center, and perform k-means clustering according to the selected initial cluster center to generate a corresponding number of clusters; calculate the average intra-cluster variation coefficient and the minimum inter-cluster variation coefficient, and then calculate the average intra-cluster variation The difference between the coefficient and the minimum inter-cluster variation coefficient, compare the difference with the set value, if the difference is less than the set value, merge the two clusters with the smallest inter-cluster variation coefficient; until the difference is greater than or equal to If the value is set, the clustering result will be output.

Description

technical field [0001] The invention relates to a clustering method, system and medium for automatically confirming the number of clusters based on the coefficient of variation. Background technique [0002] With the rapid development of information technology, many industries, such as commerce, enterprises, scientific research institutions and government departments, have accumulated massive amounts of data stored in different forms. These massive amounts of data often contain various useful information. It is difficult to obtain these information only relying on the query and retrieval mechanism of the database and statistical methods, so data mining technology is also developing rapidly. Clustering analysis technology is an important research field in data mining and has been widely used in many applications. , including pattern recognition, data analysis, image processing, and market research. [0003] Clustering analysis technology is an unsupervised learning method, i...

AI Technical Summary

Problems solved by technology

[0004] In order to solve the deficiencies of the prior art, the present invention provides a clustering method, system and medium for automatically confirming the number of clusters based on the coefficient of variation, which solves the defects of the traditional k-means++ clustering algorithm manually setting the number of clusters and improper selection of initial centroids , using the concept of variation coefficient and density index to improve the division-based k-means++ clustering algorithm, without manually setting the number of clusters, it also ensures the accuracy of the clustering results;

Images