multilayer.extraction: multilayer.extraction
In jdwilson4/MultilayerExtraction:

Function that identifies statistically significant vertex-layer communities in multilayer networks.

1 2	multilayer.extraction(adjacency, seed = 123, min.score = 0, prop.sample = 0.05, directed = c(FALSE, TRUE))

`adjacency:`	a list object whose tth entry is an adjacency matrix representing the tth layer of a multilayer network.
`seed:`	seed for reproducibility. The initial neighborhoods that act as seeds for the multilayer extraction algorithm are random in this algorithm; hence, a seed will need to be set for reproducible results. Default is 123.
`min.score:`	the minimum score allowable for an extracted community. Default is 0.
`prop.sample:`	the proportion of vertices one would like to search over for initialization. Example: prop.sample = 0.05 specifies that one will obtain 0.05 * n randomly selected vertex neighborhoods for initialization, where n = number of nodes in each layer. Default is 0.05.

A MultilayerCommunity object, which is a list containing the following objects

Community.List: a list of vertex-layer communities extracted from the algorithm
Diagnostics: the diagnostics associated with each extracted community. This is a summary of each community, and includes for each level of overlap parameter Beta the mean score, and the total number of communities. This is used for determining the overall number of communities in a multilayer network.

James D. Wilson

Wilson, James D., Palowitch, John, Bhamidi, Shankar, and Nobel, Andrew B. (2017) "Significance based extraction in multilayer networks with heterogeneous community structure." Journal of Machine Learning Research

jdwilson4/MultilayerExtraction documentation built on May 18, 2019, 11:39 p.m.