gov.sandia.cognition.learning.algorithm.nearest

Interface KNearestNeighbor<InputType,OutputType>

Type Parameters:

InputType - Type of data upon which the KNearestNeighbor operates, something like Vector, Double, or String

OutputType - Output of the evaluator, like Matrix, Double, String

All Superinterfaces:

java.lang.Cloneable, CloneableSerializable, Evaluator<InputType,OutputType>, NearestNeighbor<InputType,OutputType>, java.io.Serializable

All Known Implementing Classes:

AbstractKNearestNeighbor, KNearestNeighborExhaustive, KNearestNeighborExhaustive.Learner, KNearestNeighborKDTree, KNearestNeighborKDTree.Learner

@PublicationReference(author="Wikipedia",
                      title="k-nearest neighbor algorithm",
                      type=WebPage,
                      year=2008,
                      url="http://en.wikipedia.org/wiki/K-nearest_neighbor_algorithm")
public interface KNearestNeighbor<InputType,OutputType>
extends NearestNeighbor<InputType,OutputType>

A generic k-nearest-neighbor classifier. This classifier simply looks at the nearest "k" neighbors to a point and returns the average of them. Thus, the learner is trivial, but the lookups (evaluations) are expensive. For example, with k=1, then k-nearest-neighbor simply returns the nearest data point to a given input. And so forth.

Since:

3.0

Author:

Kevin R. Dixon

Field Summary

Fields

Modifier and Type	Field and Description
static int	DEFAULT_K The default value for k is 1.

Method Summary

Modifier and Type	Method and Description
Summarizer<? super OutputType,? extends OutputType>	getAverager() Getter for averager.
int	getK() Getter for k
void	setAverager(Summarizer<? super OutputType,? extends OutputType> averager) Setter for averager.
void	setK(int k) Setter for k

Methods inherited from interface gov.sandia.cognition.learning.algorithm.nearest.NearestNeighbor

add, getData, getDivergenceFunction

Methods inherited from interface gov.sandia.cognition.evaluator.Evaluator

evaluate

Methods inherited from interface gov.sandia.cognition.util.CloneableSerializable

clone

Field Detail

DEFAULT_K

static final int DEFAULT_K

The default value for k is 1.

See Also:

Constant Field Values

Method Detail

getAverager

Summarizer<? super OutputType,? extends OutputType> getAverager()

Getter for averager.

Returns:

Creates a single object from a collection of data

setAverager

void setAverager(Summarizer<? super OutputType,? extends OutputType> averager)

Setter for averager.

Parameters:

averager - Creates a single object from a collection of data

getK

int getK()

Getter for k

Returns:

Number of neighbors to consider, must be greater than zero

setK

void setK(int k)

Setter for k

Parameters:

k - Number of neighbors to consider, must be greater than zero