co.vote {advclust}R Documentation

Consensus / Ensemble Fuzzy Clustering: Voting

Description

Provide consensus / ensemble fuzzy clustering with voting method. Several option for voting step provided

Usage

co.vote(object, method)

Arguments

object

a fuzzycluster_list object

method

voting step that used to combine the partition ("sum","borda","product")

Details

Consensus clustering is method for combine several result of clustering into one robust result. This method used to overcome unstability of cluster result.

This function perform consensus clustering with voting approach. Voting approach look the domination of membership with several algorithm like sum rule, product rule and borda rule.

The differences of that method are how to combine several membership. Sum rule use Sum operation. Product rule use Product operation, and Borda use Borda count algorithm.

Value

Fuzzy Consensus Object

Slots

member

membership matrix

hard.label

hard.label

method.consensus

method of consensus

References

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

Examples

fuzzy.CM(iris[,1:4],K=2,m=2,max.iteration=20,threshold=1e-3,RandomNumber=1234)->cl1
fuzzy.GK(iris[,1:4],K=2,m=2,max.iteration=20,threshold=1e-3,RandomNumber=1234)->cl2
fuzzy.CM(iris[,1:4],K=2,m=2,max.iteration=20,threshold=1e-3,RandomNumber=1234)->cl3
c_fuzzycluster(cl1,cl2,cl3)->CL
co.vote(CL,"borda")

[Package advclust version 0.4 Index]