Plot networks

plot_networks() draws a network from the columns created by the previous steps of the pipeline: coordinates (layout_networks()), colours (color_networks()) and cluster labels (name_clusters() and prepare_label_networks()). Edges are drawn as light arcs whose width follows their weight, nodes as points whose size follows node_size_column, and each cluster gets a label at its centre of gravity.

It is meant as a quick, ready-made figure. For a figure to publish, print_plot_code = TRUE prints the ggraph code, which can be copied and adapted.

Usage

plot_networks(
  graphs,
  x = "x",
  y = "y",
  cluster_label_column = "cluster_label",
  node_size_column = "node_size",
  color_column = "color",
  color_networks = FALSE,
  color = NULL,
  max_arcs = 20000,
  print_plot_code = FALSE
)

Arguments

graphs

A tbl_graph or a list of tbl_graphs.

x, y

Character scalars, the node columns with the coordinates ("x" and "y" by default).

cluster_label_column

Character scalar, the node column with the labels to display: "cluster_label" (name_clusters()) by default, or a cluster column such as cluster_leiden.

node_size_column

Character scalar, the numeric node column giving the size of nodes: "node_size" by default (see compute_size in build_network()). If NULL or absent, all nodes have the same size.

color_column

Character scalar, the column with the colours of nodes and edges ("color" by default, from color_networks()).

color_networks

Logical. If TRUE, color_networks() is run first, with cluster_label_column as the categories.

color

With color_networks = TRUE, the colours passed to color_networks().

max_arcs

Whole number. Edges are drawn as arcs whose width follows their weight, unless a graph has more than max_arcs edges (20,000 by default): they are then drawn as straight lines of constant width, which is much faster (on a network of 140,000 edges, about 2 seconds and 1 GB of memory to draw, against 100 seconds and 11 GB with arcs). Set it to Inf to always draw arcs.

print_plot_code

Logical. If TRUE, prints the ggplot2/ggraph code of the plot.

Value

For a tbl_graph, a ggplot object. For a list of graphs, a named list of ggplot objects, one per graph (e.g. per time window).

Details

Missing pieces are filled in: labels are positioned with prepare_label_networks() when label_x is absent, and edges without a weight column get a weight of 1. Labels are drawn with ggrepel, which must be installed.

Examples

# \donttest{
network <- build_network(
  nodes = subset(Nodes_stagflation, source_type == "Stagflation"),
  directed_edges = Ref_stagflation,
  source_id = "source_id",
  target_id = "target_id",
  method = "similarity",
  edges_threshold = 1
)
#> ℹ Coupling (`biblionetwork::biblio_coupling()`) with "similarity": ties
#>   weighted by the idf-weighted cosine of the two lists of targets, after Shen
#>   et al. (2019), between source_id entities sharing target_id entities.
#> ℹ Two source_id are linked when they share at least 1 target_id
#>   (`edges_threshold`).
#> ℹ Nodes without any edge are removed (`keep_singleton = FALSE`).
network <- add_clusters(
  network,
  clustering_method = "leiden",
  objective_function = "modularity",
  seed = 123,
  verbose = FALSE
)
set.seed(1)
network <- layout_networks(network, node_id = "source_id", layout = "fr")
network <- color_networks(network, column_to_color = "cluster_leiden")
if (requireNamespace("ggrepel", quietly = TRUE)) {
  plot_networks(
    network,
    cluster_label_column = "cluster_leiden",
    node_size_column = NULL
  )
}

# }