Fuzzy clustering algorithms and their cluster validity

European Journal of Operational ResearchPublished 1 July 1982

Marc Roubens

DOI10.1016/0377-2217(82)90228-4

Source URL

Citations224

SJR quartileQ1

SJR score2.24

SNIP2.62

TL;DR

The global optimal solution is shown to be difficult to obtain and an alternative iterative procedure is presented which is easily implemented and converges to a local optimum.

Abstract

A non metric clustering algorithm based on a fuzzy objective function which reflects proximity based on some global dissimilarity measure is proposed. The global optimal solution is shown to be difficult to obtain and an alternative iterative procedure is presented. This procedure is easily implemented and converges to a local optimum. Some validity functionals which measure the effectiveness with which cluster structure has been identified are compared in relation with the iterative procedures described in the paper.

Keywords

Computer Science

Journal of CyberneticsA Fuzzy Relative of the ISODATA Process and Its Use in Detecting Compact Well-Separated Clusters

6,548 Citations1973J. C. Dunn

Two fuzzy versions of the k-means optimal, least squared error partitioning problem are formulated for finite subsets X of a general inner product space; in both cases, the extremizing solutions are shown to be fixed points of a certain operator T on the class of fuzzy, k-partitions of X, and simple iteration of T provides an algorithm which has the descent property relative to the least squarederror criterion function.

Journal of CyberneticsCluster Validity with Fuzzy Sets

1,214 Citations1973James C. Bezdek

This paper uses membership function matrices associated with fuzzy c-partitions of X, together with their values in the Euclidean (matrix) norm, to formulate an a posteriori method for evaluating algorithmically suggested clusterings of X.

Fuzzy Sets and SystemsPattern classification problems and fuzzy sets

209 Citations1978Marc Roubens

A unified presentation of classical clustering algorithms is proposed both for the hard and fuzzy pattern classification problems, and two coefficients that measure the “degree of non-fuzziness” of the partition are proposed.

Fuzzy Sets and SystemsCluster validity for fuzzy clustering algorithms

166 Citations1981Michael P. Windham

The proportion exponent is introduced as a measure of the validity of the clustering obtained for a data set using a fuzzy clustering algorithm and its use as a validity functional is illustrated with four numerical examples and its effectiveness compared to other validity functionals.

International Journal of Parallel ProgrammingStatistical parameters of cluster validity functionals

36 Citations1980James C. Bezdek, Michael P. Windham +1 more

The theoretical mean and variance of two related cluster validity functionals, namely, the classification entropy and partition coefficients associated with fuzzyc-partitions of finite data sets, are derived to provide a basis for statistical tests of fuzzy cluster validity.