cluster_eval: Calculate the rand-index of two label assignments
In xiangli2pro/hbcm: HBCM model for cluster analysis

matchLabel

R Documentation

Calculate the rand-index of two label assignments

Description

'matchLabel( )' calculate the rand-index and adjusted rand-index of estimated label assignment compared to true label assignment, which can be used to evaluate the performance of the estimated label. The metric takes value between 0 and 1, and higher value indicates better performance.

'calRand( )' calculates the rand-index from a confusion matrix.

'rSpecc( )' a customized spectral clustering model.

Usage

matchLabel(reference, label)

calRand(confmat)

rSpecc(x, centers, iter.max = 100, nstart = 10)

Arguments

`reference`	true label assignment.
`label`	estimated label assignment.
`confmat`	a 2-dimensional confusion matrix.
`x`	numeric matrix of data.
`centers`	the number of clusters.
`iter.max`	the maximum number of iterations allowed.
`nstart`	how many random sets in the kmeans step should be chosen? default is 10.

Value

`Rand`	rand-index
`adjRand`	adjusted rand-index
`.Data`	A vector of integers indicating the cluster to which each point is allocated.
`size`	The number of points in each cluster.
`totss`	The total sum of squares.
`withinss`	Vector of within-cluster sum of squares, one component per cluster.
`tot.withinss`	Total within-cluster sum of squares, i.e. 'sum(withinss)'.
`betweenss`	The between-cluster sum of squares, i.e. 'totss-tot.withinss'.