Robust centroids using fuzzy clustering with feature partitions

Fuzzy c-means with feature partitions uses a generalized metric on feature subsets to increase centroid robustness. Each feature partition may use a unique metric and is weighted for relevance. This method is demonstrated on synthetic and real datasets.

Saved in:
Bibliographic Details
Published inPattern recognition letters Vol. 26; no. 8; pp. 1039 - 1046
Main Authors Alexiuk, Mark D., Pizzi, Nicolino J.
Format Journal Article
LanguageEnglish
Published Elsevier B.V 01.06.2005
Subjects
Functional magnetic resonance imaging
Fuzzy c-means
Noise
Functional magnetic resonance imaging
Noise
Fuzzy c-means
Online AccessGet full text

Cover

More Information
Summary:Fuzzy c-means with feature partitions uses a generalized metric on feature subsets to increase centroid robustness. Each feature partition may use a unique metric and is weighted for relevance. This method is demonstrated on synthetic and real datasets.
ISSN:0167-8655
1872-7344
DOI:10.1016/j.patrec.2004.09.055