Tune the number of clusters

Usage

tune_k_clusters(x, ...)

# Default S3 method
tune_k_clusters(
  x,
  return_cluster = FALSE,
  plot_k = FALSE,
  method = "Spectral",
  tune_method = "silhouette",
  gap_w = "uniform",
  prox = FALSE,
  ...
)

# S3 method for class 'mrf3'
tune_k_clusters(
  x,
  return_cluster = FALSE,
  plot_k = FALSE,
  method = "Spectral",
  tune_method = "silhouette",
  gap_w = "uniform",
  prox = FALSE,
  ...
)

spectral_cl(x, k_tune = seq(2, 12, by = 1), gap_w = "uniform", d = NULL, ...)

pam_cl(
  x,
  k_tune = seq(2, 9, by = 1),
  diss = TRUE,
  tune_method = "silhouette",
  ...
)

Arguments

x: A proximity/similarity matrix (or embedding input) for clustering.
...: Additional arguments passed to lower-level clustering functions.
return_cluster: Logical; return the complete tuning result instead of only the selected number of clusters.
plot_k: Logical; draw the tuning criterion over the evaluated values of k.
method: Clustering backend: "PAM" or "Spectral".
tune_method: Tuning criterion for PAM ("silhouette" or "ratio"). The "ratio" criterion requires dissimilarities scaled to [0, 1].
gap_w: Weighting scheme for spectral eigengap ("uniform" or "log").
prox: Logical; whether x is a proximity matrix (converted to a dissimilarity matrix for PAM).
k_tune: Candidate cluster counts to evaluate.
d: Optional degree vector used by spectral clustering.
diss: logical flag: if TRUE (default for dist or dissimilarity objects), then x will be considered as a dissimilarity matrix. If FALSE, then x will be considered as a matrix of observations by variables.

Value

The selected number of clusters, or the complete clustering result when return_cluster = TRUE.