tna (version 0.4.0)

communities: Community Detection for Transition Networks

Description

This function detects communities within the transition networks (represented by the tna object). It uses various algorithms to find communities in the graph representation of transitions and returns a list of communities for each cluster or a specified cluster. If multiple transition matrices exist, the function iterates over each cluster in the tna object to find communities using different algorithms. The function uses the igraph package to convert the transition matrices into graphs and then applies community detection algorithms (e.g., Walktrap, Fast Greedy, Label Propagation, Infomap, Edge Betweenness, Leading Eigenvector, and Spin Glass).

Usage

communities(x, ...)
# S3 method for tna
communities(x, methods, gamma = 1, ...)
# S3 method for group_tna
communities(x, methods, ...)

Value

An object of class tna_communities which is a list with an element for each cluster containing:

• counts: A list with the number of communities found by each algorithm.

• assignments: A data.frame where each row corresponds to a node and each column to a community detection algorithm, with color-coded community assignments.

If x is a group_tna object, a group_tna_communities object is returned instead, which is a list of tna_communities objects.

Arguments

x

A tna or a group_tna object.

...

Ignored.

methods

A character vector of community detection algorithms to apply to the network. The supported options are:

• "walktrap": A community detection method using short random walks.

• "fast_greedy": A method based on modularity optimization.

• "label_prop": A method that uses label propagation.

• "infomap": A method that uses information flow to detect communities.

• "edge_betweenness": A method that uses edge betweenness to find communities.

• "leading_eigen": A method using the leading eigenvector of the modularity matrix.

• "spinglass": A method based on the spinglass model.

If not provided, all methods are applied.

gamma

A numeric value depicting a parameter that affects the behavior of certain algorithms like the Spin Glass method. Defaults to 1.

See Also

Pattern-finding functions plot.tna_communities(), plot.tna_comparison()

Cluster-related functions bootstrap(), centralities(), cliques(), deprune(), estimate_cs(), group_model(), hist.group_tna(), mmm_stats(), plot.group_tna(), plot.group_tna_centralities(), plot.group_tna_cliques(), plot.group_tna_communities(), plot.group_tna_stability(), plot_compare.group_tna(), plot_mosaic.group_tna(), plot_mosaic.tna_data(), print.group_tna(), print.group_tna_bootstrap(), print.group_tna_centralities(), print.group_tna_cliques(), print.group_tna_communities(), print.group_tna_stability(), print.summary.group_tna(), print.summary.group_tna_bootstrap(), prune(), pruning_details(), rename_groups(), reprune(), summary.group_tna(), summary.group_tna_bootstrap()

Examples

model <- tna(group_regulation)
comm <- communities(model)