ExactSampling-package: ExactSampling: risk evaluation using exact resampling methods...

In ExactSampling: ExactSampling: risk evaluation using exact resampling methods for the k Nearest Neighbor algorithm.

Description

ExactSampling contains functions that estimate the risk associated to a kNN prediction function (either in regression or binary classification) based on resampling method (leave-p-out cross-validation or bootstrap). It also contain a function knn.search to obtain the list of k nearest neighbors for each point in the sample, and a function to compute the empirical risk of a kNN prediction function.

Details

Package:	ExactSampling
Type:	Package
Version:	1.0
Date:	2013-01-22
License:	GPL(>=2)

Author(s)

KAI LI Maintainer: Koskas Michel<[e-mail]>

References

Celisse, A.and Mary-Huard, T. (2011) Exact Cross-Validation for kNN and applications to passive and active learning in classification. Journal de la SFdS, 152, 3.

See Also

knn.emp for the empirical estimation of the risk, knn.boot for an exact bootstrap estimation, knn.cv for the cross-validated estimation for the kNN algorithm, and knn.search to obtain the k nearest neighbors.

ExactSampling documentation built on May 2, 2019, 6:08 p.m.