KNNClassifier

java.lang.Object
- keel.Algorithms.RST_Learning.KNNClassifier

```
public class KNNClassifier
extends java.lang.Object
```
File: KNNClassifier.java A KNN classifier with the capabilities of selecting instances and features. For efficiency, employs unsquared euclidean distance.

Since:

JDK1.5

Version:

1.0

Author:

Written by Joaquín Derrac (University of Granada) 20/04/2010

Constructor Summary

Constructors
Constructor and Description

KNNClassifier()

Constructors
Constructor and Description
`KNNClassifier()`

Method Summary

All Methods Static Methods Concrete Methods
Modifier and Type	Method and Description
`static double`	`accuracy()` Estimates the LVO (Leave-one-out) accuracy of the classifier over the training data.
`static int`	`classifyNewInstance(double[] example)` Classifies a new example
`static int`	`classifyTrainingInstance(int index)` Classifies a training example
`static double`	`computeFSReduction()` Computes reduction rates over features
`static double`	`computeISReduction()` Computes reduction rates over instances
`static int[]`	`getFS()` Get a vector with the features currently selected
`static int[]`	`getIS()` Get a vector with the instances currently selected
`static java.lang.String`	`printFS()` Returns a string representation of the features selection vector
`static void`	`setAllFeatures()` On the features selector vector, sets the all the features to 1 (selected)
`static void`	`setAllInstances()` On the instance selector vector, sets the all the instances to 1 (selected)
`static void`	`setClasses(int value)` Sets the number of classes in the data
`static void`	`setData(double[][] newData)` Loads the training data into the classifier
`static void`	`setFeatures(int[] selected)` Sets the vector of features selected
`static void`	`setInstances(int[] selected)` Sets the vector of instances selected
`static void`	`setK(int value)` Sets the K value
`static void`	`setOutput(int[] newOutput)` Loads the training output into the classifier

Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

- Constructor Detail
  - KNNClassifier
```
public KNNClassifier()
```
- Method Detail
  - setClasses
```
public static void setClasses(int value)
```
    Sets the number of classes in the data
    
    Parameters:
    
    value - Number of classes
  - setK
```
public static void setK(int value)
```
    Sets the K value
    
    Parameters:
    
    value - K value
  - setData
```
public static void setData(double[][] newData)
```
    Loads the training data into the classifier
    
    Parameters:
    
    newData - Data represented with continuous values
  - setOutput
```
public static void setOutput(int[] newOutput)
```
    Loads the training output into the classifier
    
    Parameters:
    
    newOutput - Output attribute of the training data
  - setInstances
```
public static void setInstances(int[] selected)
```
    Sets the vector of instances selected
    
    Parameters:
    
    selected - Vector of instances selected
  - setFeatures
```
public static void setFeatures(int[] selected)
```
    Sets the vector of features selected
    
    Parameters:
    
    selected - Vector of features selected
  - setAllInstances
```
public static void setAllInstances()
```
    On the instance selector vector, sets the all the instances to 1 (selected)
  - setAllFeatures
```
public static void setAllFeatures()
```
    On the features selector vector, sets the all the features to 1 (selected)
  - accuracy
```
public static double accuracy()
```
    Estimates the LVO (Leave-one-out) accuracy of the classifier over the training data.
    
    Returns:
    
    Accuracy estimated
  - classifyNewInstance
```
public static int classifyNewInstance(double[] example)
```
    Classifies a new example
    
    Parameters:
    
    example - Example to classify
    
    Returns:
    
    Class of the example
  - classifyTrainingInstance
```
public static int classifyTrainingInstance(int index)
```
    Classifies a training example
    
    Parameters:
    
    index - Training example to classify
    
    Returns:
    
    Class of the example
  - computeISReduction
```
public static double computeISReduction()
```
    Computes reduction rates over instances
    
    Returns:
    
    Reduction rate
  - computeFSReduction
```
public static double computeFSReduction()
```
    Computes reduction rates over features
    
    Returns:
    
    Reduction rate
  - getIS
```
public static int[] getIS()
```
    Get a vector with the instances currently selected
    
    Returns:
    
    A vector with the instances currently selected
  - getFS
```
public static int[] getFS()
```
    Get a vector with the features currently selected
    
    Returns:
    
    A vector with the features currently selected
  - printFS
```
public static java.lang.String printFS()
```
    Returns a string representation of the features selection vector
    
    Returns:
    
    a string representation of the features selection vector

Class KNNClassifier

Constructor Summary

Method Summary

Methods inherited from class java.lang.Object

Constructor Detail

KNNClassifier

Method Detail

setClasses

setK

setData

setOutput

setInstances

setFeatures

setAllInstances

setAllFeatures

accuracy

classifyNewInstance

classifyTrainingInstance

computeISReduction

computeFSReduction

getIS

getFS

printFS