0

I have a dataset like following :

        link      from_node  to_node
 1   412300044     100923     100185
 2   412300047     100190     100198
 3   412300057     100197     100198
 4   412300058     100198     100199
 5   412300076     100190     100199
...     ...          ...        ...

I will use 'link' variable as an edge, and 'from_node' and 'to_node' as vertices. No weighted value currently. I decided this, but have no idea what can I do further. I searched internet about making edge betweenness figure but there were very few...I would be very helpful if you help me.

hogu
  • 1
  • 3
  • 1
    Please read this: https://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example and include some attempted code. You might also want to read `igraph` tutorials. https://igraph.org/r/#rdoc1 https://rstudio-pubs-static.s3.amazonaws.com/74248_3bd99f966ed94a91b36d39d8f21e3dc3.html – John Polo Nov 27 '22 at 13:16
  • 1
    Please, include dput(dataframe) into your question. – clp Nov 27 '22 at 14:08

2 Answers2

2

We don't have your data set, but here's one that approximates it:

set.seed(1234)

df <- data.frame(link      = sample(30, 10) + 412300040,
                 from_node = sample(10, 10, T) + 100184,
                 to_node   = sample(10, 10, T) + 100184)

df
#>         link from_node to_node
#> 1  412300068    100188  100189
#> 2  412300056    100186  100192
#> 3  412300066    100191  100188
#> 4  412300062    100190  100192
#> 5  412300045    100194  100187
#> 6  412300052    100190  100188
#> 7  412300055    100188  100194
#> 8  412300049    100192  100189
#> 9  412300070    100188  100186
#> 10 412300046    100188  100192

If you simply want to create a graph and calculate edge betweenness using igraph, you can do:

library(igraph)

ig <- graph_from_data_frame(df[c(3, 2, 1)])
edge_betweenness(ig)
#> [1] 2.5 2.0 6.0 1.5 4.0 4.5 6.0 2.5 3.0 2.0

To plot with edges weighted according to betweenness, you could do:

plot(ig, edge.width = edge_betweenness(ig), edge.label = df$link)

enter image description here

However, I think if you want to plot graphs, you should check out the tidygraph and ggraph packages, which allow easy manipulation of graphs, and highly customizable plots of graphs respectively. Here's how you could plot your graph with edges weighted according to betweenness:

library(tidygraph)
library(ggraph)

as_tbl_graph(df[c(3, 2, 1)]) %>%
  activate(edges) %>%
  mutate(betweenness = centrality_edge_betweenness()) %>%
  ggraph() +
  geom_edge_diagonal(aes(edge_width = betweenness, label = link), 
                     angle_calc = "along", alpha = 0.5, vjust = -1,
                     color = "lightblue") +
  geom_node_circle(aes(r = 0.2), fill = "lightblue", color = "navy") +
  geom_node_text(aes(label = name)) +
  coord_equal() +
  theme_graph()

enter image description here

Allan Cameron
  • 147,086
  • 7
  • 49
  • 87
1

Try this.

require(igraph)
df <- data.frame(
   edge      = c(412300044,  412300047, 412300057, 412300058, 412300076),
   from_node = c(100923, 100190 , 100197 , 100198 , 100190),
   to_node   = c(100185, 100198 , 100198 , 100199 , 100199)
)

# reorder, rename columns
df2 <- data.frame(to_node=df[,2], from_node=df[,3], label=df[,1])

 g <- graph_from_data_frame(df2, directed=TRUE)
plot(g)

Note that the command help(graph_from_data_frame) can help you further.

clp
  • 1,098
  • 5
  • 11
  • This doesn't seem to include the edge betweenness (though to be fair, it's not entirely clear exactly what the OP wanted here) – Allan Cameron Nov 27 '22 at 14:03
  • Good point. I interpreted "edge betweenness" as edge link, which may not be intended. But including edge betweenness is doable given the examples. – clp Nov 27 '22 at 14:22