Skip to contents

Turn a list of networks into the table of an alluvial plot

Source: R/networks_to_alluv.R
networks_to_alluv.Rd

[Experimental]

An alluvial plot shows the evolution of dynamic clusters: each time window is a column, each cluster a block whose height is its share of the nodes of the window, and the flows between blocks follow the nodes that stay in the network from one window to the next (Rosvall and Bergstrom 2010) . networks_to_alluv() stacks the node tables of a list of networks into the long table that such a plot needs (one row per node and window) and adds summary statistics on the clusters.

Usage

networks_to_alluv(
  graphs,
  cluster_id,
  node_id,
  summary_cluster_stats = TRUE,
  keep_color = TRUE,
  color_column = "color",
  keep_cluster_label = TRUE,
  cluster_label_column = "cluster_label"
)

Arguments

graphs

A list of tbl_graphs with dynamic clusters (see merge_dynamic_clusters()).

cluster_id

Character scalar, the node column with the dynamic clusters (e.g. "dynamic_cluster_leiden").

node_id

Character scalar, the node column with a stable identifier; it is the "alluvium" that the flows follow.

summary_cluster_stats

Logical. If TRUE (default), adds the statistics described in the return value, useful to keep only the main clusters in the plot.

keep_color

Logical. If TRUE (default), keeps color_column.

color_column

Character scalar, the colour column ("color", from color_networks()).

keep_cluster_label

Logical. If TRUE (default), keeps cluster_label_column.

cluster_label_column

Character scalar, the label column ("cluster_label", from name_clusters()).

Value

A data.table with one row per node and window: window (the name of the graph in the list), node_id, the dynamic cluster, the colour and label columns if kept, y_alluv (\(1/n_t\), \(n_t\) being the number of nodes of the window, so that each window has a total height of 1), and, with summary_cluster_stats = TRUE:

  • share_cluster_alluv: the share of all rows that belong to the cluster, in per cent;

  • share_cluster_window: the share of the window's nodes in the cluster, in per cent;

  • share_cluster_max: the maximum of share_cluster_window over the windows;

  • length_cluster: the number of windows where the cluster exists.

Details

The alluvial method (following clusters over time with networks_to_alluv(), minimize_crossing_alluvial() and plot_alluvial()) was initiated by Alexandre Truc.

References

Rosvall M, Bergstrom CT (2010). “Mapping Change in Large Networks.” PLoS ONE, 5(1), e8694. doi:10.1371/journal.pone.0008694 .

See also

minimize_crossing_alluvial(), prepare_label_alluvial(), color_alluvial() and plot_alluvial().

Examples

networks <- build_dynamic_networks(
  nodes = subset(Nodes_stagflation, source_type == "Stagflation"),
  directed_edges = Ref_stagflation,
  source_id = "source_id",
  target_id = "target_id",
  time_variable = "source_year",
  time_window = 10,
  method = "similarity",
  edges_threshold = 1,
  verbose = FALSE
)
#> Warning: The last window is shorter than the others.
#> ℹ The data end in "2013", but the last window ends in "2014".
networks <- add_clusters(
  networks,
  clustering_method = "leiden",
  objective_function = "modularity",
  seed = 123,
  verbose = FALSE
)
networks <- merge_dynamic_clusters(
  networks,
  cluster_id = "cluster_leiden",
  node_id = "source_id",
  method = "majority", threshold_similarity = 0.51,
  similarity_type = "partial"
)
#> ℹ Linking clusters across windows with the "majority" rule: a cluster inherits
#>   the identifier of a past cluster when more than 0.51 of the nodes of each go
#>   to / come from the other (double majority). The comparison counts the nodes
#>   present in both windows only (`similarity_type` = "partial").
#> Warning: 3 pairs of consecutive windows share no node: their clusters cannot be matched.
#> ℹ Clusters are matched through the nodes they share. In a network of documents
#>   (coupling), each node has one date: the windows must overlap (`window_step`
#>   smaller than `time_window` in `build_dynamic_networks()`). If windows already
#>   overlap, the nodes may have changed completely, or the threshold may leave a
#>   window without edges.

alluv_dt <- networks_to_alluv(
  networks,
  cluster_id = "dynamic_cluster_leiden",
  node_id = "source_id"
)
#> ℹ No color column in the graphs: no colour kept (see `color_alluvial()`).
#> ℹ No cluster_label column in the graphs: no label kept (see `name_clusters()`).
alluv_dt[1:5]
#>    dynamic_cluster_leiden    window  source_id share_cluster_alluv
#>                    <char>    <fctr>     <char>               <num>
#> 1:                   cl_1 1975-1984    1172921               11.11
#> 2:                   cl_1 1975-1984   14371905               11.11
#> 3:                   cl_1 1975-1984    2901822               11.11
#> 4:                   cl_1 1975-1984 1111111128               11.11
#> 5:                   cl_1 1975-1984    1184127               11.11
#>    share_cluster_window share_cluster_max length_cluster  y_alluv
#>                   <num>             <num>          <int>    <num>
#> 1:                   25                25              1 0.015625
#> 2:                   25                25              1 0.015625
#> 3:                   25                25              1 0.015625
#> 4:                   25                25              1 0.015625
#> 5:                   25                25              1 0.015625