Title :
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
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;
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
DOI :
10.1109/ISUMA.1995.527768
