spec_clust {nett} | R Documentation |
Perform spectral clustering (with regularization) to estimate communities
spec_clust(
A,
K,
type = "lap",
tau = 0.25,
nstart = 20,
niter = 10,
ignore_first_col = FALSE
)
A |
Adjacency matrix (n x n) |
K |
Number of communities |
type |
("lap" | "adj" | "adj2") Whether to use Laplacian or adjacency-based spectral clustering |
tau |
Regularization parameter for the Laplacian |
nstart |
argument from function 'kmeans' |
niter |
argument from function 'kmeans' |
ignore_first_col |
whether to ignore the first eigen vector when doing spectral clustering |
A label vector of size n x 1 with elements in 1,2,...,K