NormMI: Calculate Normalized Mutual Information

In SeojinBang/MKKC: Multiple Kernel K-means Clustering

Description

Calculate Normalized Mutual Information measuring the mutual dependence between two clustering labels.

Usage

NormMI(x, y)

MI(x, y)

Entropy(x)

Arguments

x	A vector of class labels.
y	A vector of class labels to be compared with x. The length of y should be the same as x.

Value

NormMI returns Normalized Mutual Information between two label vectors x and y.

MI returns Mutual Information between two label vectors.

Entropy returns Entropy of the label vector x.

References

\insertRef

strehl2002clusterMKKC

See Also

AdjRandIndex, Purity

Examples

# true label
x <- rep(1:3, each = 10)

# predicted label
y <- sample(x)

# compare x and y
AdjRandIndex(x, y)
NormMI(x, y)
Purity(x, y)

SeojinBang/MKKC documentation built on Sept. 18, 2019, 1:42 p.m.