Soft Voting Cluster Ensemble

Description

This function used to perform Soft Voting Cluster Ensemble.

Usage

1
2
soft.vote.ensemble(data, seed, method = "FCM", K = 2, m = 2, gamma = 0,
  rho = rep(1, K), threshold = 10^-5, max.iteration = 100, core)

Arguments

data

data frame nxp

seed

number of ensemble

method

fuzzy clustering method that will be used ("FCM" or "GK")

K

specific number of cluster (must be >1)

m

fuzzifier / degree of fuzziness

gamma

parameter of Gustafson Kessel Clustering

rho

parameter of volume clustering in Gustafson Kessel Clustering

threshold

threshold of convergence

max.iteration

maximum iteration to convergence

core

number of core that used for parallelization

Details

Soft vote cluster ensemble used to stabilize the result of cluster analysis. It can be define combine several result of clustering to be one robust result.

The simple method of ensemble is voting method, vote label that resulted and use maximum number of voting as partition. For fuzzy clustering, voting method use membership matrix. This function implemented voting method with sum rule approach. For standarize the label, this function use hungary algorithm for optimal labelization.

Value

func.obj objective function that calculated.

U matrix n x K consist fuzzy membership matrix

V matrix K x p consist fuzzy centroid

D matrix n x K consist distance of data to centroid that calculated

Clust.desc cluster description (dataset with additional column of cluster label)

seeding list of random number that used as seeding

Call call argument

References

Sevillano, X., Alias, F., & Socoro, J. C. (2013). Posisional and Confidence voting-based Consensus Function For Fuzzy Cluster Ensemble. Fuzzy Sets and System, 1-40.

Examples

1
2
#library(RcmdrPlugin.FuzzyClust)
#soft.vote.ensemble(iris[1:50,1:4],seed=2,method="FCM",core=1,max.iteration=20,threshold=10^-3)->Cl