visualize overlapping communities in graph by any of the python or R modules

Question

How can I visualize communities if there are overlapping communities in the graph? I can use any module in python (networkx, igraph, matplotlib, etc.) or R.

For example, information on nodes, edges, and the nodes in each community is given as follows. Note that node G spans two communities.

list_nodes = ['A', 'B', 'C', 'D','E','F','G','H','I','J']
tuple_edges = [('A','B'),('A','C'),('A','D'),('B','C'),('B','D'), ('C','D'),('C','E'),
              ('E','F'),('E','G'),('F','G'),('G','H'),
              ('G','I'), ('G','J'),('H','I'),('H','J'),('I','J'),]
list_communities = [['A', 'B', 'C', 'D'],['E','F','G'],['G', 'H','I','J']]

I would like a plot that visualizes the community as shown below.

In networkx, it is possible to colour-code each node like this post, but this method is not suitable when communities overlap.

In igraph, communities can be visualised using the community extraction method included in the package, as described in this post. However, in my case I want to define communities using the list of nodes contained in each community.

Question is similar to `Plot communities with igraph`, https://stackoverflow.com/questions/26913419/plot-communities-with-igraph — clp, Nov 29 '22 at 19:28

ThomasIsCoding · Accepted Answer · 2022-11-30T10:25:31.833

3

Below is an option for igraph within R.

I think you may have to annotate the community info manually (see grp below) and then use it when plotting, e.g.,

g <- graph_from_data_frame(df, directed = FALSE)
grp <- lapply(
  groups(cluster_edge_betweenness(g)),
  function(x) {
    c(
      x,
      names(which(colSums(distances(g, x) == 1) > 1))
    )
  }
)
plot(g, mark.groups = grp)

or using max_cliques with minimal size 3 (thank @clp's comment)

g <- graph_from_data_frame(df, directed = FALSE)
grp <- max_cliques(g, min = 3)
plot(g, mark.groups = grp)

Data

df <- data.frame(
  from = c("A", "A", "A", "B", "B", "C", "C", "E", "E", "F", "G", "G", "G", "H", "H", "I"),
  to = c("B", "C", "D", "C", "D", "D", "E", "F", "G", "G", "H", "I", "J", "I", "J", "J")
)

edited Nov 30 '22 at 10:25

answered Nov 29 '22 at 09:42

ThomasIsCoding

96,636
9
24
81

1

The community structure in this question can also be re-engineered by `grp <- max_cliques(g, min=3)`. – clp Nov 29 '22 at 19:39
@clp Yes, exactly, that simplifies a lot! Thanks for the advice! – ThomasIsCoding Nov 29 '22 at 19:53
@ThomasIsCoding, new to graph networks, any good tutorials or book I can use to learn quickly and use same on a project? – wwnde Dec 10 '22 at 03:17
1

@wwnde Perhaps starting from any textbook of graph theory – ThomasIsCoding Dec 10 '22 at 21:27

visualize overlapping communities in graph by any of the python or R modules

1 Answers1

Data