Release Time:2019-03-11  Hits:

Indexed by: Journal Article

Date of Publication: 2016-03-01

Journal: Journal of Computational and Theoretical Nanoscience

Included Journals: Scopus、EI

Volume: 13

Issue: 3

Page Number: 2157-2161

ISSN: 15461955

Abstract: A novel cluster validity index is proposed for the validation of partitions for object data. The cluster validity index ranges from the creation of the subclasses produced by the fuzzy c-means algorithm up to their fusion which implies proximity tree according to the largest separation criterion. Testing of the proposed index and four previously formulated indices on well-known data sets shows the superior effectiveness and reliability of the proposed index in comparison with other indices. Copyright ? 2016 American Scientific Publishers. All rights reserved.