On robustifying the C-Means algorithms

Author

Kim, Jongwoo ; Krishnapuram, Raghu ; Davé, Rajesh

Author_Institution

Dept. of Electr. & Comput. Eng., Missouri Univ., Columbia, MO, USA

fYear

1995

Firstpage

630

Lastpage

635

Abstract

The Hard and Fuzzy C-Means algorithms are commonly used in many applications. However, they are highly sensitive to noise and outliers. We reformulate the Hard and Fuzzy C-Means algorithms and combine them with a robust estimator called the Least Trimmed Squares to produce robust versions of these algorithms. To find the optimum trimming ratio of the data set and to eliminate the noise from the data set, we develop an unsupervised algorithm based on a cluster validity measure. We illustrate the robustness of these algorithm with examples

Keywords

fuzzy set theory; least mean squares methods; numerical stability; statistical analysis; Fuzzy C-Means algorithm; Hard C-Means algorithm; Least Trimmed Squares; cluster validity measure; noise; optimum trimming ratio; outliers; robust estimator; robustness; unsupervised algorithm; Clustering algorithms; Clustering methods; Industrial engineering; Least squares methods; Noise measurement; Noise robustness; Parameter estimation; Pollution measurement; Prototypes; Statistics;

fLanguage

English

Publisher

ieee

Conference_Titel

Uncertainty Modeling and Analysis, 1995, and Annual Conference of the North American Fuzzy Information Processing Society. Proceedings of ISUMA - NAFIPS '95., Third International Symposium on

Conference_Location

College Park, MD

Print_ISBN

0-8186-7126-2

Type

conf

DOI

10.1109/ISUMA.1995.527768

Filename

527768