Setup a SpectralLaplaceConfig object for spectral clustering with an
exponential (Laplacian) kernel, via the 'kernlab' package. The exponential
kernel decays more slowly than the Gaussian of setup_SpectralRBF and so
keeps more weight on distant pairs.
setup_SpectralLaplace(
k = 3L,
sigma = NULL,
iterations = 200L,
nystrom = NULL,
features = NULL
)SpectralLaplaceConfig object.
Integer [2, Inf): Number of clusters. Must be no more than the number of cases.
Optional Numeric (0, Inf): Inverse kernel width. Unset uses 1.
Integer [1, Inf): Maximum number of k-means iterations run on the spectral embedding.
Optional NystromConfig: Approximate the affinity matrix from a sample of the cases, from setup_Nystrom. NULL decomposes it in full.
Optional Character vector: Names of at least 2 distinct
feature columns to cluster on. NULL clusters on all numeric features.
EDG
Argument names are rtemis' own. They map to kernlab::specc() as: k ->
centers, sigma -> kpar = list(sigma) with kernel = "laplacedot",
nystrom -> nystrom.red and nystrom.sample; iterations keeps its
name.
Ng AY, Jordan MI, Weiss Y (2001). On Spectral Clustering: Analysis and an Algorithm. Advances in Neural Information Processing Systems, 14, 849-856.
setup_SpectralLaplace(k = 3L, sigma = 0.5)
Run the code above in your browser using DataLab