Please use this identifier to cite or link to this item: https://hdl.handle.net/20.500.14279/1570
Title: Factor analysis latent subspace modeling and robust fuzzy clustering using t-distributions
Authors: Chatzis, Sotirios P. 
Varvarigou, Theodora 
Major Field of Science: Engineering and Technology
Field Category: Electrical Engineering - Electronic Engineering - Information Engineering
Keywords: Fuzzy systems;Factor analysis;Gaussian distribution;Expectation-maximization algorithms
Issue Date: Jun-2009
Source: IEEE Transactions on Fuzzy Systems, 2009, vol. 17, no. 3, pp. 505-517
Volume: 17
Issue: 3
Start page: 505
End page: 517
Journal: IEEE Transactions on Fuzzy Systems 
Abstract: Factor analysis is a latent subspace model commonly used for local dimensionality reduction tasks. Fuzzy c-means (FCM) type fuzzy clustering approaches are closely related to Gaussian mixture models (GMMs), and expectation - maximization (EM) like algorithms have been employed in fuzzy clustering with regularized objective functions. Student's t-mixture models (SMMs) have been proposed recently as an alternative to GMMs, resolving their outlier vulnerability problems. In this paper, we propose a novel FCM-type fuzzy clustering scheme providing two significant benefits when compared with the existing approaches. First, it provides a well-established observation space dimensionality reduction framework for fuzzy clustering algorithms based on factor analysis, allowing concurrent performance of fuzzy clustering and, within each cluster, local dimensionality reduction. Second, it exploits the outlier tolerance advantages of SMMs to provide a novel, soundly founded, nonheuristic, robust fuzzy clustering framework by introducing the effective means to incorporate the explicit assumption about Student's t-distributed data into the fuzzy clustering procedure. This way, the proposed model yields a significant performance increase for the fuzzy clustering algorithm, as we experimentally demonstrate
URI: https://hdl.handle.net/20.500.14279/1570
ISSN: 19410034
DOI: 10.1109/TFUZZ.2008.924317
Rights: © IEEE
Type: Article
Affiliation : National Technical University Of Athens 
Publication Type: Peer Reviewed
Appears in Collections:Άρθρα/Articles

CORE Recommender
Show full item record

SCOPUSTM   
Citations

38
checked on Nov 8, 2023

WEB OF SCIENCETM
Citations 1

29
Last Week
0
Last month
1
checked on Oct 29, 2023

Page view(s)

471
Last Week
2
Last month
2
checked on Nov 21, 2024

Google ScholarTM

Check

Altmetric


Items in KTISIS are protected by copyright, with all rights reserved, unless otherwise indicated.