select {adjclust} | R Documentation |
Clustering selection from a chac object with the slope heuristic or the broken stick heuristic
select(x, type = c("capushe", "bstick"), k.max = NULL, graph = FALSE, pct = 0.15)
x |
an object of class 'chac' |
type |
model selection approach between slope heuristic
( |
k.max |
maximum number of clusters that can be selected. Default to
|
graph |
logical. Whether the diagnostic plot for the capushe selection
is displayed or not. Default to |
pct |
minimum percentage of points for the plateau selection in
capushe selection. See |
The function returns the clustering selected by the slope heuristic,
as implemented in the R package capushe
.
Baudy, J.P., Maugis, C. and Michel, B. (2012) Slope heuristics: overview and implementation. Statistics and Computing, 22(2), 355-470. MacArthur, R.H. (1957) On the relative abundance of bird species. Proceedings of the National Academy of Sciences, 43, 293-295.
## Not run: if (require("HiTC", quietly = TRUE)) { load(system.file("extdata", "hic_imr90_40_XX.rda", package = "adjclust")) res <- hicClust(hic_imr90_40_XX, log = TRUE) selected.capushe <- select(res) table(selected.capushe) selected.bs <- select(res, type = "bstick") table(selected.bs) } ## End(Not run) res <- adjClust(dist(iris[ ,1:4])) select.clust <- select(res, "bs") table(select.clust)