dist_to_knn: Given a distance matrix, find the k nearest neighbors.

In BenjaK/scanstatistics: Space-Time Anomaly Detection using Scan Statistics

Description

Given a distance matrix, calculate the k nearest neighbors of each location, including the location itself. The matrix should contain only zeros on the diagonal, and all other elements should be positive.

Usage

dist_to_knn(x, k = min(10, nrow(x)))

Arguments

x	A (square) distance matrix. Elements should be non-negative and the diagonal zeros, but this is not checked.
k	The number of nearest neighbors, counting the location itself.

Value

A matrix of integers, row i containing the k nearest neighbors of location i, including itself.

Examples

x <- matrix(c(0, 0,
              1, 0,
              2, 1,
              0, 4,
              1, 3),
            ncol = 2, byrow = TRUE)
d <- dist(x, diag = TRUE, upper = TRUE)
dist_to_knn(d, k = 3)

BenjaK/scanstatistics documentation built on May 5, 2019, 2:41 p.m.