Unsupervised data classification method based on generalized fuzzy clustering algorithm

The invention discloses an unsupervised data classification method based on a generalized fuzzy clustering algorithm. The method comprises the steps that optimal division is performed on a sample setaccording to the GFC objective function minimization principle; position and velocity values of multi...

Full description

Saved in:
Bibliographic Details
Main Authors WEN CHUANJUN, XU DINGLIANG, LIU FUYAN
Format Patent
LanguageChinese
English
Published 26.10.2018
Subjects
CALCULATING
COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
COMPUTING
COUNTING
HANDLING RECORD CARRIERS
PHYSICS
PRESENTATION OF DATA
RECOGNITION OF DATA
RECORD CARRIERS
Online AccessGet full text

Cover

More Information
Summary:The invention discloses an unsupervised data classification method based on a generalized fuzzy clustering algorithm. The method comprises the steps that optimal division is performed on a sample setaccording to the GFC objective function minimization principle; position and velocity values of multiple particles are initialized; the particle position values and sample clustering centers are correspondingly subjected to clustering center initialization; it is defined that the distance between each sample and each clustering center and fuzzy membership are in a inversely proportional relationship, so that the sample fuzzy membership is calculated; updated clustering centers are obtained according to a particle swarm optimization algorithm iteration formula; and a GFC objective function is obtained through calculation. The constructed fuzzy clustering algorithm is not limited by normalization constraint, and noise data can be effectively mined and recognized; and the constructed inversely proportional relationsh
Bibliography:Application Number: CN201810495011