homogeneity: Calculates the homogeneity of a clustering.
In laderast/Consense: Consense: a package for evaluating multiple clusterings

Description Usage Arguments Value Author(s) References Examples

Given cluster output from a suitable clustering function, calculate the Homogeneity (from Shamir, et al.) of the clusterings.

Homogeneity is a measure of the within-cluster variance. See Shamir, et al, for how it is calculated.

1	homogeneity(clustering, centers, distfunc="euclidean")

`clustering`	Cluster output from a clustering function. the output should be in matrix format with the cluster assignments being the first column in that matrix, with the actual data in remaining columns.
`centers`	Cluster Centers as returned by `ClusterCenters`.
`distfunc`	Distance function to be used. Currently only "euclidean" or "correlation" are supported.

homogeneity

An unbounded, non-negative scalar that represents the homogeneity of the clustering.

Ted Laderas (laderast@ohsu.edu)

Shamir, et al. Algorithmic Approaches to Clustering Gene Expression Data. in Current Topics in Computational Molecular Biology. MIT Press: Boston.

    ##calculate homogeneity for Cho et al's clusters
    data(chocellcycle)
    data(choclusters)
    cho <- cbind(as.numeric(choclusters), chocellcycle)
    cent <- clusterCenters(as.data.frame(cho))
    homogeneity(cho, cent)