Hi, I have a set of nodes and a dissimilarity matrix for them, as well as a csv file in which the diss matrix has been converted to [node_1, node_2, dissimilarity] format. I would like to visualize this as a graph in Euclidean space (that is, similar nodes clumped together in clusters), rather than the seriation visualization given by dissplot(). I am using Network WorkBench for my visualizations and thus want the R output to be in graphml. If I use, say, graph.data.frame(), it will read the dissimilarity column as an edge attribute rather than as distance between nodes, which is what I want. How should I go about this? Many thanks! [[alternative HTML version deleted]]
Corey Sparks
2011-Jul-09 15:50 UTC
[R] Visualizing a dissimilarity matrix in Euclidean space
Hi, I've playd with this kind of problem before, have yoiu looked at principal coordinates? You can use the cmdscale() function in R to take the eigenstructure of your distance matrix and plot the differences in low dimensional space, it can be very instructive as to the dissimilarity between your notwork nodes. CS ----- Corey Sparks, PhD Assistant Professor Department of Demography and Organization Studies University of Texas at San Antonio 501 West Durango Blvd Monterey Building 2.270C San Antonio, TX 78207 210-458-3166 corey.sparks 'at' utsa.edu https://rowdyspace.utsa.edu/users/ozd504/www/index.htm -- View this message in context: http://r.789695.n4.nabble.com/Visualizing-a-dissimilarity-matrix-in-Euclidean-space-tp3654720p3656354.html Sent from the R help mailing list archive at Nabble.com.