Last data update: 2014.03.03
|
R: Support Criterion
rknnSupport | R Documentation |
Support Criterion
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<lishengqiao@yahoo.com>
Results
|