affinity_matrix: Generate a symmetric affinity matrix based on a distance...

In ANF: Affinity Network Fusion for Complex Patient Clustering

Description

Generate a symmetric affinity matrix based on a distance matrix using 'local' Gaussian kernel

Usage

affinity_matrix(D, k, alpha = 1/6, beta = 1/6)

Arguments

D	distance matrix (need to be a square and non-negative matrix)
k	the number of k-nearest neighbors
alpha	coefficient for local diameters. Default value: 1/6. This default value should work for most cases.
beta	coefficient for pair-wise distance. Default value: 1/6. This default value should work for most cases.

Value

an affinity matrix

Examples

D = matrix(runif(400), nrow=20)
A = affinity_matrix(D, 5)

Example output

ANF documentation built on Nov. 8, 2020, 7:51 p.m.