Cluster-based Similarity Partitioning Algorithm (CSPA)
Performs hierarchical clustering on a stack of consensus matrices to obtain consensus class labels.
CSPA(E, k)
E |
is an array of clustering results. |
k |
number of clusters |
cluster assignments for the consensus class
Derek Chiu
Other consensus functions:
LCA(),
LCE(),
k_modes(),
majority_voting()
data(hgsc)
dat <- hgsc[1:100, 1:50]
x <- consensus_cluster(dat, nk = 4, reps = 4, algorithms = c("hc", "diana"),
progress = FALSE)
CSPA(x, k = 4)Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.