On robustifying the C-means algorithms

Jongwoo Kim; Raghu Krishnapuram; Rajesh Dave

On robustifying the C-means algorithms

Jongwoo Kim, Raghu Krishnapuram, Rajesh Dave

Research output: Contribution to conference › Paper › peer-review

Abstract

The Hard and Fuzzy C-Means algorithms are commonly used in many applications. However, they are highly sensitive to noise and outliers. In this paper, 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.

Original language	English (US)
Pages	630-635
Number of pages	6
State	Published - 1995
Externally published	Yes
Event	Proceedings of the 3rd International Symposium on Uncertainty Modeling and Analysis and Annual Conference of the North American Fuzzy Information Processing Society, (ISUMA - NAFIPS'95) - College Park, MD, USA Duration: Sep 17 1995 → Sep 20 1995

Other

Other	Proceedings of the 3rd International Symposium on Uncertainty Modeling and Analysis and Annual Conference of the North American Fuzzy Information Processing Society, (ISUMA - NAFIPS'95)
City	College Park, MD, USA
Period	9/17/95 → 9/20/95

All Science Journal Classification (ASJC) codes

General Computer Science
General Mathematics

Cite this

@conference{fff022a9546f4acf94a64f78bd2b537e,

title = "On robustifying the C-means algorithms",

abstract = "The Hard and Fuzzy C-Means algorithms are commonly used in many applications. However, they are highly sensitive to noise and outliers. In this paper, 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.",

author = "Jongwoo Kim and Raghu Krishnapuram and Rajesh Dave",

year = "1995",

language = "English (US)",

pages = "630--635",

note = "Proceedings of the 3rd International Symposium on Uncertainty Modeling and Analysis and Annual Conference of the North American Fuzzy Information Processing Society, (ISUMA - NAFIPS'95) ; Conference date: 17-09-1995 Through 20-09-1995",

}

On robustifying the C-means algorithms. / Kim, Jongwoo; Krishnapuram, Raghu; Dave, Rajesh.
1995. 630-635 Paper presented at Proceedings of the 3rd International Symposium on Uncertainty Modeling and Analysis and Annual Conference of the North American Fuzzy Information Processing Society, (ISUMA - NAFIPS'95), College Park, MD, USA.

Research output: Contribution to conference › Paper › peer-review

TY - CONF

T1 - On robustifying the C-means algorithms

AU - Kim, Jongwoo

AU - Krishnapuram, Raghu

AU - Dave, Rajesh

PY - 1995

Y1 - 1995

N2 - The Hard and Fuzzy C-Means algorithms are commonly used in many applications. However, they are highly sensitive to noise and outliers. In this paper, 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.

AB - The Hard and Fuzzy C-Means algorithms are commonly used in many applications. However, they are highly sensitive to noise and outliers. In this paper, 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.

UR - http://www.scopus.com/inward/record.url?scp=38449103931&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=38449103931&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:38449103931

SP - 630

EP - 635

T2 - Proceedings of the 3rd International Symposium on Uncertainty Modeling and Analysis and Annual Conference of the North American Fuzzy Information Processing Society, (ISUMA - NAFIPS'95)

Y2 - 17 September 1995 through 20 September 1995

ER -

On robustifying the C-means algorithms

Abstract

Other

All Science Journal Classification (ASJC) codes

Other files and links

Fingerprint

Cite this