org.neuroph.eval.classification

Class ClassificationMetrics

java.lang.Object

org.neuroph.eval.classification.ClassificationMetrics

public final class ClassificationMetrics
extends java.lang.Object

Container class for all metrics which use confusion matrix for their computation Based on: http://java-ml.sourceforge.net/api/0.1.7/net/sf/javaml/classification/evaluation/PerformanceMeasure.html http://sourceforge.net/p/java-ml/java-ml-code/ci/a25ddde7c3677da44e47a643f88e32e2c8bbc32f/tree/net/sf/javaml/classification/evaluation/PerformanceMeasure.java http://en.wikipedia.org/wiki/Matthews_correlation_coefficient

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	ClassificationMetrics.Stats

Constructor Summary

Constructors

Constructor and Description
ClassificationMetrics(int truePositive, int trueNegative, int falsePositive, int falseNegative) Constructs a new measure using arguments TODO: add class to which measure corresponds?

Method Summary

Modifier and Type	Method and Description
static ClassificationMetrics.Stats	average(ClassificationMetrics[] results)
static ClassificationMetrics[]	createFromMatrix(ConfusionMatrix confusionMatrix)
double	getAccuracy() Calculate and return classification accuracy measure.
double	getBalancedClassificationRate()
java.lang.String	getClassLabel() Returns class label for
double	getErrorRate() A number of wrong predictions made divided by the total number of predictions made.
double	getFalseDiscoveryRate()
double	getFalseNegativeRate()
double	getFalsePositiveRate()
double	getFMeasure() Calculates F-score for beta equal to 1.
double	getFMeasure(int beta) Returns the F-score.
double	getMatthewsCorrelationCoefficient()
double	getPrecision() Calculate and return classification precision measure.
double	getQ9()
double	getRecall()
double	getSensitivity() Calculate and return classification sensitivity (recall, true positive rate) number of correctly classified positive examples divided by the total number of actual positive examples
double	getSpecificity() Specifity , true negative rate
double	getTotal() Returns total number of classifications.
void	setClassLabel(java.lang.String classLabel)
java.lang.String	toString()

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, wait, wait, wait

Constructor Detail

ClassificationMetrics

public ClassificationMetrics(int truePositive,
                             int trueNegative,
                             int falsePositive,
                             int falseNegative)

Constructs a new measure using arguments TODO: add class to which measure corresponds?

Parameters:

truePositive -

trueNegative -

falsePositive -

falseNegative -

Method Detail

getClassLabel

public java.lang.String getClassLabel()

Returns class label for

Returns:

class labels

setClassLabel

public void setClassLabel(java.lang.String classLabel)

getAccuracy

public double getAccuracy()

Calculate and return classification accuracy measure. A number of correct predictions made divided by the total number of predictions made a = ( tp + tn ) / n

Returns:

classification accuracy

getErrorRate

public double getErrorRate()

A number of wrong predictions made divided by the total number of predictions made. Also error = 1 - accuracy

Returns:

getPrecision

public double getPrecision()

Calculate and return classification precision measure. A number of correctly classified positive examples divided by the total number of examples that are classified as positive Also known as positive predictive value PPV

Returns:

classification precision measure

getSensitivity

public double getSensitivity()

Calculate and return classification sensitivity (recall, true positive rate) number of correctly classified positive examples divided by the total number of actual positive examples

Returns:

getRecall

public double getRecall()

getSpecificity

public double getSpecificity()

Specifity , true negative rate

getTotal

public double getTotal()

Returns total number of classifications.

Returns:

total number of classifications

getFalsePositiveRate

public double getFalsePositiveRate()

getFalseNegativeRate

public double getFalseNegativeRate()

getFalseDiscoveryRate

public double getFalseDiscoveryRate()

getMatthewsCorrelationCoefficient

public double getMatthewsCorrelationCoefficient()

getFMeasure

public double getFMeasure()

Calculates F-score for beta equal to 1.

Returns:

f-score

getFMeasure

public double getFMeasure(int beta)

Returns the F-score. When recall and precision are zero, this method will return 0.

Parameters:

beta -

Returns:

f-score

getQ9

public double getQ9()

getBalancedClassificationRate

public double getBalancedClassificationRate()

toString

public java.lang.String toString()

Overrides:

toString in class java.lang.Object

createFromMatrix

public static ClassificationMetrics[] createFromMatrix(ConfusionMatrix confusionMatrix)

average

public static ClassificationMetrics.Stats average(ClassificationMetrics[] results)

Parameters:

results - list of different metric results computed on different sets of data

Returns:

average metrics computed different MetricResults