Comparison new quality indices for fuzzy c-means clustering

Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21630%2F19%3A00349413" target="_blank" >RIV/68407700:21630/19:00349413 - isvavai.cz</a>
Result on the web
<a href="http://evlm.stuba.sk/APLIMAT/indexe.htm" target="_blank" >http://evlm.stuba.sk/APLIMAT/indexe.htm</a>
DOI - Digital Object Identifier
—

Result language
angličtina
Original language name
Comparison new quality indices for fuzzy c-means clustering
Original language description
Evaluating fuzzy clustering results is still an actual problem. In the evaluation of fuzzy clustering results, it is necessary, in the case of quantitative variables, to compute multiple indices because there is no universal index to determine the correct number of clusters. For determining the correct number of clusters, it is more accurate/precise to use indices based not only on membership function, but indices based on membership function and data set. In most cases, those types of indices were able to recognize the correct number of clusters.
Czech name
—
Czech description
—

Project
—
Continuities
I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Publication year
2019
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection
18th Conference on Applied Mathematics (APLIMAT 2019)
ISBN
9781510882140
ISSN
—
e-ISSN
—
Number of pages
7
Pages from-to
1002-1008
Publisher name
Slovak University of Technology in Bratislava
Place of publication
Bratislava
Event location
Bratislava
Event date
Feb 5, 2019
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
—

Similar results(10)