R Graphical Manual

Browse All

Last data update: 2014.03.03

R: Perform k-Nearest Neighbors Lookup Using a Search Tree

knnLookup

R Documentation

Perform k-Nearest Neighbors Lookup Using a Search Tree

Description

This function performs fast k-Nearest Neighbors lookup on a SearchTree object

Usage

knnLookup(tree, newx, newy, newdat, columns = 1:2, k = 5)

Arguments

`tree`	An object which inherits from the `SearchTree` S4 class.
`newx`	Numeric. Vector of x values for the points to look up neighbors for.
`newy`	Numeric. Vector of x values for the points to look up neighbors for.
`newdat`	Matrix or data.frame. Data containing x and y values of the points to look up neighbors for. Ignored if `x` and `y` are specified.
`columns`	Numeric. Columns x and y values can be found in within `newdat`.
`k`	Numeric. Number of neighbors to find for each point.

Value

The return value is an integer matrix indicating the indices in the original data used to create treE where the nearest neighbors were found. Row indicates the indice of the new point, while column indicates the order of the k neighbors.

Note

No defined order is specified for exact ties in distance.

Author(s)

Gabriel Becker

Examples

x = rnorm(100)
y = rnorm(100)
tree = createTree(cbind(x,y))
newx = c(0, .5)
newy = c(.5, 0)
inds = knnLookup(tree, newx, newy, k=7)

ch = rep(1, times=100)
ch[inds[1:7]] = 3
ch[inds[8:14]] = 5
cls = rep("black", times=100)
cls[inds[1:7]] = "red"
cls[inds[8:14]] ="blue"

plot(x,y, pch=ch, col = cls)
abline(v=newx[1], h = newy[1] , col="red")
abline(v=newx[2], h = newy[2], col = "blue")

Results


R version 3.3.1 (2016-06-21) -- "Bug in Your Hair"
Copyright (C) 2016 The R Foundation for Statistical Computing
Platform: x86_64-pc-linux-gnu (64-bit)

R is free software and comes with ABSOLUTELY NO WARRANTY.
You are welcome to redistribute it under certain conditions.
Type 'license()' or 'licence()' for distribution details.

R is a collaborative project with many contributors.
Type 'contributors()' for more information and
'citation()' on how to cite R or R packages in publications.

Type 'demo()' for some demos, 'help()' for on-line help, or
'help.start()' for an HTML browser interface to help.
Type 'q()' to quit R.

> library(SearchTrees)
> png(filename="/home/ddbj/snapshot/RGM3/R_CC/result/SearchTrees/knnLookup.Rd_%03d_medium.png", width=480, height=480)
> ### Name: knnLookup
> ### Title: Perform k-Nearest Neighbors Lookup Using a Search Tree
> ### Aliases: knnLookup 'knnLookup, QuadTree-method'
> ### Keywords: knn lookup neighbors
> 
> ### ** Examples
> 
> x = rnorm(100)
> y = rnorm(100)
> tree = createTree(cbind(x,y))
> newx = c(0, .5)
> newy = c(.5, 0)
> inds = knnLookup(tree, newx, newy, k=7)
> 
> ch = rep(1, times=100)
> ch[inds[1:7]] = 3
> ch[inds[8:14]] = 5
> cls = rep("black", times=100)
> cls[inds[1:7]] = "red"
> cls[inds[8:14]] ="blue"
> 
> plot(x,y, pch=ch, col = cls)
> abline(v=newx[1], h = newy[1] , col="red")
> abline(v=newx[2], h = newy[2], col = "blue")
> 
> 
> 
> 
> 
> dev.off()
null device 
          1 
>