diva.sketch.classification

Class KNNClassifier

java.lang.Object

diva.sketch.classification.KNNClassifier

All Implemented Interfaces:

Classifier, TrainableClassifier

public class KNNClassifier
extends Object
implements TrainableClassifier

A K Nearest Neighbor classifier compares a given example (feature set) to the training set and make its prediction based on the majority match in the top K candidates.

Version:

$Revision: 1.2 $

Author:

Heloise Hse (hwawen@eecs.berkeley.edu)

Field Summary

Fields

Modifier and Type	Field and Description
static double[]	f1
static double[]	f2
static double[]	f3
static double[]	f4
static double[]	f5
static double[]	f6
static double[]	f7
static double[]	f8
static double[]	f9

Constructor Summary

Constructors

Constructor and Description
KNNClassifier()
KNNClassifier(int k) Create a K nearest neighbor classifier in which K is set to specified number.

Method Summary

Modifier and Type	Method and Description
Classification	classify(FeatureSet s) Compare the given example (features) to the examples in the training set and save the top K closest matches in ascending distance values.
void	clear() Clear all results of previous trainings (presumably so that this classifier can be trained again from scratch).
boolean	isIncremental() Return true.
static void	main(String[] argv)
void	setK(int k) Set the value of K.
void	train(TrainingSet s) Store the given training set which will be used during classification.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Detail

f1

public static double[] f1

f2

public static double[] f2

f3

public static double[] f3

f4

public static double[] f4

f5

public static double[] f5

f6

public static double[] f6

f7

public static double[] f7

f8

public static double[] f8

f9

public static double[] f9

Constructor Detail

KNNClassifier

public KNNClassifier()

KNNClassifier

public KNNClassifier(int k)

Create a K nearest neighbor classifier in which K is set to specified number.

Method Detail

setK

public void setK(int k)

Set the value of K. The classifier finds the closest K matches.

train

public void train(TrainingSet s)
           throws ClassifierException

Store the given training set which will be used during classification. If a training set already exist, this method simply adds the example in the given training set (s) to the existing set. This also means that if an example in "s" already exists in the current training set, it'll be added again.

Specified by:

train in interface TrainableClassifier

Throws:

ClassifierException

isIncremental

public boolean isIncremental()

Return true. This classifier is incremental. It supports multiple calls to "train".

Specified by:

isIncremental in interface TrainableClassifier

clear

public void clear()

Clear all results of previous trainings (presumably so that this classifier can be trained again from scratch). This simply sets the current training set to null.

Specified by:

clear in interface TrainableClassifier

classify

public Classification classify(FeatureSet s)
                        throws ClassifierException

Compare the given example (features) to the examples in the training set and save the top K closest matches in ascending distance values.

Specified by:

classify in interface Classifier

Throws:

ClassifierException

main

public static void main(String[] argv)