netrankr-package: netrankr: An R package for centrality and partial rankings in...

In netrankr: Analyzing Partial Rankings in Networks

netrankr: An R package for centrality and partial rankings in networks

Description

netrankr provides several functions to analyze partial rankings for network centrality. The main focus lies on methods that do not necessarily rely on indices like degree, betweenness or closeness. However, the package also provides more than 20 indices, which can be constructed via a Rstudio addin.

The package follows the philosophy, that centrality can be decomposed in a series of micro steps. Starting from a network, indirect_relations can be derived which can either be aggregated into an index with aggregate_positions, or alternatively turned into a partial ranking with positional_dominance. The partial ranking can then be further analyzed with exact_rank_prob, to obtain probabilistic centrality rankings.

Details

Some features of the package are:

• Working with the neighborhood inclusion preorder. This forms the bases for any centrality analysis on undirected and unweighted graphs. More details can be found in the dedicated vignette: vignette("neighborhood_inclusion",package = "netrankr")

• Constructing graphs with a unique centrality ranking. This class of graphs, known as threshold graphs, can be used to benchmark centrality indices, since they only allow for one ranking of the nodes. For more details consult the vignette: vignette("threshold_graph",package = "netrankr")

• Probabilistic centrality. Why apply a handful of indices and choosing the one that fits best, when it is possible to analyze all centrality rankings at once? The package includes several function to calculate rank probabilities of nodes in a network. These include expected ranks and relative rank probabilities (how likely is it that a node is more central than another?) Consult vignette("probabilistic_cent",package = "netrankr") for more info.

The package provides several additional vignettes that explain the functionality of netrankr and its conceptual ideas. See browseVignettes(package = 'netrankr')

Author(s)

Maintainer: David Schoch [e-mail] (ORCID)

Other contributors:

• Julian Müller [e-mail] [contributor]

See Also

Useful links:

• https://github.com/schochastics/netrankr/

• https://schochastics.github.io/netrankr/

• Report bugs at https://github.com/schochastics/netrankr/issues

netrankr documentation built on May 29, 2024, 4:19 a.m.