estimate_k: estimate_k: estimate K using the eigengap or multimodality...

In Spectrum: Fast Adaptive Spectral Clustering for Single and Multi-View Data

Description

This function will try to estimate K given a similarity matrix. Generally the maximum eigengap is preferred, but on some data examining the distribution of the eigenvectors as in the multimodality gap heuristic may be beneficial.

Usage

estimate_k(A2, maxk = 10, showplots = TRUE)

Arguments

A2	Data frame or matrix: a similarity matrix
maxk	Numerical value: maximum number of K to be considered
showplots	Character value: whether to show the plot on the screen

Value

A data frame containing the eigenvalues and dip-test statistics of the eigenvectors of the graph Laplacian

Examples

k_test <- estimate_k(missl[[1]])

Example output

egap optimal K: 8

Spectrum documentation built on Feb. 10, 2020, 9:07 a.m.