Hi, here is a simple example that helps understand it: g <- graph.star(100, mode="undirected", center=1) V(g)$type <- c(FALSE, rep(TRUE, 99)) bipartite.projection(g)
# $proj1 # IGRAPH U--B 1 0 -- Star # + attr: name (g/c), mode (g/c), center (g/n), type (v/x) # # $proj2 # IGRAPH U-WB 99 4851 -- Star # + attr: name (g/c), mode (g/c), center (g/n), type (v/x), weight (e/n) Gabor On Fri, Nov 15, 2013 at 5:50 PM, <[email protected]> wrote: > Hello, > > I have been scratching my head in trying to understand the result of a > one-mode projection of a bipartite graph. > > This is the bipartite graph > > > graph_bi > IGRAPH DN-B 154625 448384 -- > > and after bipartite.projection(graph_bi) these are the two projections > > > graph_onemode$proj1 > IGRAPH UNWB 79186 13259012 -- > > > > graph_onemode$proj2 > IGRAPH UNWB 75439 50375826 -- > > > Now, my intuition doesn't help here. Simply, How is possible that the number > of edges in the two projections increases by 29 and 112 times respectively? > > I am probably missing something here but as I understand the one-mode > projection the number of edges should actually decrease. A-B and B-C (where > A and C are type 1 nodes and B is a type 2 node) should be projected to A-C > thus halving the number of edges... > > _______________________________________________ > igraph-help mailing list > [email protected] > https://lists.nongnu.org/mailman/listinfo/igraph-help _______________________________________________ igraph-help mailing list [email protected] https://lists.nongnu.org/mailman/listinfo/igraph-help
