FKM.ent: Fuzzy k-means with entropy regularization

Description

Performs the fuzzy k-means clustering algorithm with entropy regularization.

Usage

FKM.ent (X, k, ent, RS, stand, startU, conv, maxit)

Arguments

Matrix or data.frame

Number of clusters (default: 2)

ent

Degree of fuzzy entropy (default: 1)

Number of (random) starts (default: 1)

stand

Standardization: if stand=1, the clustering algorithm is run using standardized data (default: no standardization)

startU

Rational starting point for the membership degree matrix U (default: no rational start)

conv

Convergence criterion (default: 1e-9)

maxit

Maximum number of iterations (default: 1e+6)

Value

Object of class fclust, which is a list with the following components:
UMembership degree matrix
HPrototype matrix
FArray containing the covariance matrices of all the clusters (NULL for FKM.ent)
clusMatrix containing the indices of the clusters where the objects are assigned (column 1) and the associated membership degrees (column 2)
medoidVector containing the indices of the medoid objects (NULL for FKM.ent)
valueVector containing the loss function values for the RS starts
cputVector containing the computational times (user times) for the RS starts
iterVector containing the numbers of iterations for the RS starts
kNumber of clusters
mParameter of fuzziness (NULL for FKM.ent)
entDegree of fuzzy entropy
vpVolume parameter (NULL for FKM.ent)
deltaNoise distance (NULL for FKM.ent)
standStandardization (Yes if stand=1, No if stand=0)
XcaData used in the clustering algorithm (standardized data if stand=1)
XRaw data
callMatched call

References

Li R., Mukaidono M., 1995. A maximum entropy approach to fuzzy clustering. Proceedings of the Fourth IEEE Conference on Fuzzy Systems (FUZZ-IEEE/IFES '95), pp. 2227-2232. Li R., Mukaidono M., 1999. Gaussian clustering method based on maximum-fuzzy-entropy interpretation. Fuzzy Sets and Systems, 102, 253-258.

Examples

Run this code

data(Mc)
for (j in 2:(ncol(Mc)-1))
Mc[,j]=Mc[,j]/Mc[,1]
Mc=Mc[,-1]
## It may take more than a few seconds
clust=FKM.ent(Mc[,1:(ncol(Mc)-1)],k=6,ent=3,RS=10,stand=1)