Robust clustering methods: A unified view
Document Type
Article
Publication Date
12-1-1997
Abstract
Clustering methods need to be robust if they are to be useful in practice. In this paper, we analyze several popular robust clustering methods and show that they have much in common. We also establish a connection between fuzzy set theory and robust statistics and point out the similarities between robust clustering methods and statistical methods such as the weighted least-squares (LS) technique, the M estimator, the minimum volume ellipsoid (MVE) algorithm, cooperative robust estimation (CRE), minimization of probability of randomness (MINPRAN), and the epsilon contamination model. By gleaning the common principles upon which the methods proposed in the literature are based, we arrive at a unified view of robust clustering methods. We define several general concepts that are useful in robust clustering, state the robust clustering problem in terms of the defined concepts, and propose generic algorithms and guidelines for clustering noisy data. We also discuss why the generalized Hough transform is a suboptimal solution to the robust clustering problem. © 1997 IEEE.
Identifier
0031139176 (Scopus)
Publication Title
IEEE Transactions on Fuzzy Systems
External Full Text Location
https://doi.org/10.1109/91.580801
ISSN
10636706
First Page
270
Last Page
293
Issue
2
Volume
5
Grant
N00014-96-1-0439
Fund Ref
Office of Naval Research
Recommended Citation
Davé, Rajesh N. and Krishnapuram, Raghu, "Robust clustering methods: A unified view" (1997). Faculty Publications. 16677.
https://digitalcommons.njit.edu/fac_pubs/16677
