Cluster management method

The invention provides a cluster management method, which relates to the technical field of integrated clustering, and comprises the following steps: extracting a corresponding partition main cluster based on partition characteristics of all partitions; on the basis of main features and secondary fe...

Full description

Saved in:
Bibliographic Details
Main Authors QI GANG, LI FANG, WANG YUBO, HAN GUOQUAN
Format Patent
LanguageChinese
English
Published 29.03.2024
Subjects
CALCULATING
COMPUTING
COUNTING
ELECTRIC DIGITAL DATA PROCESSING
PHYSICS
Online AccessGet full text

Cover

More Information
Summary:The invention provides a cluster management method, which relates to the technical field of integrated clustering, and comprises the following steps: extracting a corresponding partition main cluster based on partition characteristics of all partitions; on the basis of main features and secondary features of all partition main clusters, feature-associated sigma clusters are constructed; obtaining a corresponding first clustering similarity based on all processing features of every two sigma clusters; combining the two sigma clusters with the highest first clustering similarity into a new sigma cluster, and calculating the second clustering similarity of every two new sigma clusters again until the maximum third clustering similarity in the second clustering similarities is lower than a predefined threshold value; all sigma clusters corresponding to the third clustering similarity are obtained, the corresponding optimal cluster number is obtained, the number of clusters participating in the final consensus is
Bibliography:Application Number: CN202311567643