I experimented with HiveR and successfully created a hive chart, which is one of the data sets that I work with. The original data graph is a classic red tape:
However, the hive chart provides a much simpler and more intuitive look at the data:
However, the problem is that the plot occupies only a relatively small part of the canvas. I tried various tricks to crop it so that the plot occupied most of the frame, but none of them were successful. The most promising options were to write the plot in pdf or eps format, and then use pdfcrop or epstool to trim the gap, but none of them worked in practice. I suspect the plot has invisible elements that accumulate a bounding box.
Does anyone have any ideas on how to plot this plot efficiently? [I know that I can do it manually, but I will need to create a series of these stories, and I'm looking for a systematic / programmatic way to do this.]
The data structure for the hive plot can be found here .
Thanks a lot, Andrew.
Here is the code to create the plot:
set.seed(3) VERTICES <- 512 library(igraph) big.graph <- barabasi.game(VERTICES, power = 1, out.seq = sort(sample(c(1, 2, 3, 4), VERTICES, replace = TRUE, prob = c(0.1, 0.7, 0.1, 0.1))), directed = FALSE, out.pref = TRUE) library(HiveR) library(digest) V(big.graph)$name <- sapply(1:VERTICES, function(n) {digest(n, algo = "crc32", serialize = TRUE)}) big.matrix <- get.adjacency(big.graph, type= "lower", attr=NULL, names=TRUE, sparse=FALSE) colnames(big.matrix) <- V(big.graph)$name rownames(big.matrix) <- colnames(big.matrix) hive <- adj2HPD(big.matrix, axis.cols = "black") hive <- mineHPD(hive, option = "rad <- tot.edge.count") hive <- mineHPD(hive, option = "axis <- source.man.sink") summary = sumHPD(hive, chk.all = TRUE, plot.list = TRUE)