spectral_embedding: Spectral Embedding

In arthans/SpectralClustering: Spectral Clustering

Description

Project the sample on the first eigenvectors of the graph Laplacian.

Usage

spectral_embedding(adjacency, n_components = 8)

Arguments

adjacency	The adjacency matrix of the graph to embed.
n_components	The dimension of the projection subspace.

Details

The adjacency matrix is used to compute a normalized graph Laplacian whose spectrum (especially the eigenvectors associated to the smallest eigenvalues) has an interpretation in terms of minimal number of cuts necessary to split the graph into comparably sized components.

Value

The reduced samples.

References

https://en.wikipedia.org/wiki/LOBPCG

Toward the Optimal Preconditioned Eigensolver: Locally Optimal Block Preconditioned Conjugate Gradient Method Andrew V. Knyazev https://epubs.siam.org/doi/10.1137/S1064827500366124

Examples

data(affinity)

maps <- spectral_embedding(affinity, 2)

plot(maps)

arthans/SpectralClustering documentation built on Dec. 19, 2021, 4:41 a.m.