GenerateDilsNetwork: Combine multiple networks into a single weighted network.

In dils: Data-Informed Link Strength. Combine multiple-relationship networks into a single weighted network. Impute (fill-in) missing network links.

Description

Use ScalablePCA to recover optimal weights for each network, then calculate the weighted average across networks for each edge.

Usage

  GenerateDilsNetwork(x, subsample = 10000,
    n.subsamples = 1000, ignore.cols, use.cols,
    progress.bar = FALSE)

Arguments

x	data.frame, data over which to run PCA
subsample	numeric or logical, If an integer, size of each subsample. If FALSE, runs PCA on entire data set.
n.subsamples	numeric, number of subsamples.
ignore.cols	numeric, indices of columns not to include
use.cols	numeric, indices of columns to use
progress.bar	logical, if TRUE then progress in running subsamples will be shown.

Value

A list

dils	vector, named vector of component weights for first dimension of principal component analysis (see example for comparison to prcomp).
dils.edgelist	Unused columns of x bound with the DILS scores on the right. Forms an edgelist if there were two unused columns and they containted the ids for the dyads.
coefficients	named vector, weights that genereate dils by taking dot-product with network data.
weights	named vector, raw.weights scaled by standard deviations of network edges, then scaled to sum to 1.

Author(s)

Stephen R. Haptonstahl [e-mail]

References

https://github.com/shaptonstahl/

See Also

prcomp

Examples

data(iris)        # provides example data
GenerateDilsNetwork(iris, subsample=10, use.cols=1:4)
GenerateDilsNetwork(iris, subsample=10, ignore.cols=5)

dils documentation built on May 2, 2019, 8:28 a.m.