CODEWITHSUNDEEP
rigraphsocial-networkingadjacency-matrix
peterkeenan
peterkeenan

Reputation: 1

How to stop Igraph generating duplicate vertices from adjacency matrix?

I am attemping to create a weighted social network between a series of characters. I have manually created an adjacency matrix with weighted values: excel file of relationship values, 0-3. This has been linked to a separate file, in which attributes of the characters have been stored for later colouration.

My goal is to create a graph in TKplot, where vertice size is the degree centrality, and edge width is a function of their weight.

However, when using the igraph package to generate a simpler grapher, just to test the code, I noticed that I am generating additional vertices, which appear to be duplicates of the vertices (for example, see the clear and coloured squares marked "Rusty" on the right side of the image:practise igraph plot.

Here is the code I have written so far:

network_data<-read.csv('bbnmatrix2.csv',header=T, row.names=1, as.is=T)
bbnnames<-read.csv("bbnnames.csv", header=T, as.is=T)
net2 <- graph_from_incidence_matrix(network_data, directed = FALSE,
                                    mode = c("total"),
                                    multiple = FALSE,
                                    weighted = T,
                                    add.names = T)
plot(net2, vertex.label.color="black", vertex.label=bbnnames$name,
       vertex.label.cex=0.5, edge.color="black",
     vertex.color=clan, edge.width=E(net2)$weight,
     vertex.shapes=bbnnames$status).

I was expecting a plot of 67 vertices, but instead generated a plot of 134.

Upvotes: -1

Views: 129

Answers (1)

nmaclaren
nmaclaren

Reputation: 108

Are you trying to make a bipartite network? graph_from_incidence_matrix() treats your input matrix as defining a bipartite network. I think you want graph_from_adjacency_matrix() instead.

Upvotes: 0

Related Questions