rknnSupport: Support Criterion

In rknn: Random KNN Classification and Regression

Description

Compute support criterion using Random KNN classification or regression

Usage

rknnSupport(data, y, k = 1, r = 500, mtry = trunc(sqrt(ncol(data))), 
    fixed.partition = FALSE, cluster=NULL, seed = NULL)
rknnRegSupport(data, y, k = k, r = 500, mtry = trunc(sqrt(ncol(data))), 
    fixed.partition = FALSE, cluster=NULL, seed = NULL)

Arguments

data	The input dataset.
y	A vector of responses.
k	Number of nearest neighbors.
r	Number of KNNs.
mtry	Number of features to be drawn for each KNN.
fixed.partition	Logical. Use fixed partition of dynamic partition of the data into training and testing subsets for each KNN.
cluster	An object of class ‘c("SOCKcluster", "cluster")’
seed	An integer seed.

Value

A supportKNN object.

Author(s)

Shengqiao Li<[e-mail]>

rknn documentation built on May 2, 2019, 12:35 p.m.