nd.him {NetworkDistance}

HIM Distance

Description

Hamming-Ipsen-Mikhailov (HIM) combines the local Hamming edit distance and the global Ipsen-Mikhailov distance to merge information at each scale. For Ipsen-Mikhailove distance, it is provided as nd.csd in our package for consistency. Given a parameter \xi (xi), it is defined as

HIM_{\xi}(A,B)=\sqrt{H^2(A,B)+\xi\cdot IM^2(A,B)}/\sqrt{1+\xi}

where H and IM stand for Hamming and I-M distance, respectively.

Usage

nd.him(A, out.dist = TRUE, xi = 1, ntest = 10)

Arguments

A

a list of length N containing (M\times M) adjacency matrices.

out.dist

a logical; TRUE for computed distance matrix as a dist object.

xi

a parameter to control balance between two distances.

ntest

the number of searching over nd.csd parameter.

Value

a named list containing

D

an (N\times N) matrix or dist object containing pairwise distance measures.

References

Jurman G, Visintainer R, Filosi M, Riccadonna S, Furlanello C (2015). “The HIM glocal metric and kernel for network comparison and classification.” In 2015 IEEE International Conference on Data Science and Advanced Analytics (DSAA), 1–10. ISBN 978-1-4673-8272-4.

See Also

nd.hamming, nd.csd

Examples

## load example data
data(graph20)

## compute distance matrix
output = nd.him(graph20, out.dist=FALSE)

## visualize
opar = par(no.readonly=TRUE)
par(pty="s")
image(output$D[,20:1], main="two group case", axes=FALSE, col=gray(0:32/32))
par(opar)