问题
I am following the examples over here on using the "statnet" library in http://personal.psu.edu/drh20/papers/v24i09.pdf.
The first example shows how to inspect a statnet network object in R:
library(statnet)
library(network)
data("faux.magnolia.high")
fmh <- faux.magnolia.high
summary(fmh)
In the above example, it seems here that the statnet network in this example already has "node attributes".
Using the statnet library, does anyone know if there is a way to directly create a network with node attributes from a data frame?
For example, if I have some data that looks like this:
mydata <-data.frame(
"source" = c("123","124","123","125","123"),
"target" = c("126", "123", "125", "122", "111"),
"color" = c("red","red","green","blue","red"),
"food" = c("pizza","pizza","cake","pizza","cake")
)
Suppose I had a pre defined list of node attributes:
Nodes <-data.frame(
"source" = c("123","124","125","122","111", "126"),
"Country" = c("usa", "uk", "uk", "usa", "uk", "usa")
)
I tried the following code:
net = network(mydata)
But I am not sure if this has created a network with the node attributes (color and food).
I also tried this, but it did not work:
mydata <-data.frame(
"source" = c("123","124","123","125","123"), "target" = c("126", "123", "125", "122", "111"), "color" = c("red","red","green","blue","red"), "food" = c("pizza","pizza","cake","pizza","cake") )
Nodes <-data.frame(
"source" = c("123","124","125","122","111", "126"),
"Country" = c("usa", "uk", "uk", "usa", "uk", "usa")
)
net<-network(mydata[,c[1:2])
edges <- as.sociomatrix(mydata[,c(3:4)],simplify=TRUE)
nodes <- as.sociomatrix(Nodes,simplify=TRUE)
final <- as.sociomatrix(list(net,edges,nodes))
Can someone please show me how to create a network with node attributes?
source: https://rdrr.io/github/statnet/network/man/as.sociomatrix.html
Thanks
回答1:
Could this be the answer? https://igraph.org/r/doc/graph_from_data_frame.html
library(igraph)
g <- graph_from_data_frame(mydata, directed=TRUE, vertices=Nodes)
回答2:
Igraph is OK, but if you want to remain within the network package, that is statnet suite, you could do the following:
net<- as.network(mydata, matrix.type = "edgelist")
set.vertex.attribute(net, "color", as.character(mydata$color))
set.vertex.attribute(net, "food", as.character(mydata$food))
#To verify...
get.vertex.attribute(net, "color")
It seems that the set.vertex.attribute function does not accept factors, hence as.character()
If you have a lot of vertex attributes you could use "apply" to apply the set.vertex.attribute as a function over columns of a data frame with vertex attributes.
Generally, the materials (tutorials) for statnet suite from Michael Heaney were very useful to me (materials linked in the first bullet under the title "Summer Workshops"): http://michaeltheaney.com/teaching
来源：https://stackoverflow.com/questions/64852237/r-creating-a-statnet-network-with-node-attributes