gov.sandia.cognition.learning.performance.categorization

Class DefaultBinaryConfusionMatrixConfidenceInterval

java.lang.Object

gov.sandia.cognition.learning.performance.categorization.DefaultBinaryConfusionMatrixConfidenceInterval

public class DefaultBinaryConfusionMatrixConfidenceInterval
extends java.lang.Object

Puts Student-t ConfidenceIntervals on each entry of the ConfusionMatrix

Since:

2.0

Author:

Kevin R. Dixon

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	DefaultBinaryConfusionMatrixConfidenceInterval.Summary An implementation of the Summarizer interface for creating a ConfusionMatrixInterval

Constructor Summary

Constructors

Constructor and Description
DefaultBinaryConfusionMatrixConfidenceInterval(double confidence, ConfidenceInterval falsePositivesRate, ConfidenceInterval falseNegativesRate, ConfidenceInterval truePositivesRate, ConfidenceInterval trueNegativesRate) Creates a new instance of ConfusionMatrixConfidenceInterval

Method Summary

Modifier and Type	Method and Description
protected static void	checkConfidence(double confidence) Checks to make sure that confidence is between 0.0 and 1.0.
static DefaultBinaryConfusionMatrixConfidenceInterval	compute(java.util.Collection<? extends DefaultBinaryConfusionMatrix> data, double confidence) Computes the ConfidenceIntervals for the given Collection of ConfusionMatrices
double	getConfidence() Getter for confidence
ConfidenceInterval	getFalseNegativesRate() Getter for falseNegativesRate
ConfidenceInterval	getFalsePositivesRate() Getter for falsePositivesRate
ConfidenceInterval	getTrueNegativesRate() Getter for trueNegativesRate
ConfidenceInterval	getTruePositivesRate() Getter for truePositivesRate
protected void	setConfidence(double confidence) Setter for confidence
protected void	setFalseNegativesRate(ConfidenceInterval falseNegativesRate) Setter for falseNegativesRate
protected void	setFalsePositivesRate(ConfidenceInterval falsePositivesRate) Setter for falsePositivesRate
protected void	setTrueNegativesRate(ConfidenceInterval trueNegativesRate) Setter for trueNegativesRate
protected void	setTruePositivesRate(ConfidenceInterval truePositivesRate) Setter for truePositivesRate
java.lang.String	toString()

Methods inherited from class java.lang.Object

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

Constructor Detail

DefaultBinaryConfusionMatrixConfidenceInterval

public DefaultBinaryConfusionMatrixConfidenceInterval(double confidence,
                                                      ConfidenceInterval falsePositivesRate,
                                                      ConfidenceInterval falseNegativesRate,
                                                      ConfidenceInterval truePositivesRate,
                                                      ConfidenceInterval trueNegativesRate)

Creates a new instance of ConfusionMatrixConfidenceInterval

Parameters:

confidence - Confidence that the statistic is within the bound, or 1-alpha, on the interval [0,1], where confidence=0 means definitely not within the bound and confidence=1 means definitely within the bound.

falsePositivesRate - The fraction of target falses incorrectly classified as true, FalsePositives / TotalNegatives = 1 - TrueNegativesRate

falseNegativesRate - The fraction of target falses incorrectly classified as false, FalseNegatives / TotalPositives = 1 - TruePositivesRate

truePositivesRate - The fraction of target trues that were correctly classified as true, TruePositives / TotalPositives = TruePositives / (TruePositives + FalseNegatives)

trueNegativesRate - The fraction of negative targets correctly classified as false, TrueNegatives / TotalNegatives = TrueNegatives / (TrueNegatives + FalsePositives)

Method Detail

getFalsePositivesRate

public ConfidenceInterval getFalsePositivesRate()

Getter for falsePositivesRate

Returns:

The fraction of target falses incorrectly classified as true, FalsePositives / TotalNegatives = 1 - TrueNegativesRate

setFalsePositivesRate

protected void setFalsePositivesRate(ConfidenceInterval falsePositivesRate)

Setter for falsePositivesRate

Parameters:

falsePositivesRate - The fraction of target falses incorrectly classified as true, FalsePositives / TotalNegatives = 1 - TrueNegativesRate

getFalseNegativesRate

public ConfidenceInterval getFalseNegativesRate()

Getter for falseNegativesRate

Returns:

The fraction of target falses incorrectly classified as false, FalseNegatives / TotalPositives = 1 - TruePositivesRate

setFalseNegativesRate

protected void setFalseNegativesRate(ConfidenceInterval falseNegativesRate)

Setter for falseNegativesRate

Parameters:

falseNegativesRate - The fraction of target falses incorrectly classified as false, FalseNegatives / TotalPositives = 1 - TruePositivesRate

getTruePositivesRate

public ConfidenceInterval getTruePositivesRate()

Getter for truePositivesRate

Returns:

The fraction of target trues that were correctly classified as true, TruePositives / TotalPositives = TruePositives / (TruePositives + FalseNegatives)

setTruePositivesRate

protected void setTruePositivesRate(ConfidenceInterval truePositivesRate)

Setter for truePositivesRate

Parameters:

truePositivesRate - The fraction of target trues that were correctly classified as true, TruePositives / TotalPositives = TruePositives / (TruePositives + FalseNegatives)

getTrueNegativesRate

public ConfidenceInterval getTrueNegativesRate()

Getter for trueNegativesRate

Returns:

The fraction of negative targets correctly classified as false, TrueNegatives / TotalNegatives = TrueNegatives / (TrueNegatives + FalsePositives)

setTrueNegativesRate

protected void setTrueNegativesRate(ConfidenceInterval trueNegativesRate)

Setter for trueNegativesRate

Parameters:

trueNegativesRate - The fraction of negative targets correctly classified as false, TrueNegatives / TotalNegatives = TrueNegatives / (TrueNegatives + FalsePositives)

getConfidence

public double getConfidence()

Getter for confidence

Returns:

Confidence that the statistic is within the bound, or 1-alpha, on the interval [0,1], where confidence=0 means definitely not within the bound and confidence=1 means definitely within the bound.

setConfidence

protected void setConfidence(double confidence)

Setter for confidence

Parameters:

confidence - Confidence that the statistic is within the bound, or 1-alpha, on the interval [0,1], where confidence=0 means definitely not within the bound and confidence=1 means definitely within the bound.

checkConfidence

protected static void checkConfidence(double confidence)

Checks to make sure that confidence is between 0.0 and 1.0.

Parameters:

confidence - The confidence.

toString

public java.lang.String toString()

Overrides:

toString in class java.lang.Object

compute

public static DefaultBinaryConfusionMatrixConfidenceInterval compute(java.util.Collection<? extends DefaultBinaryConfusionMatrix> data,
                                                                     double confidence)

Computes the ConfidenceIntervals for the given Collection of ConfusionMatrices

Parameters:

data - Collection of ConfusionMatrices from which to compute the ConfidenceIntervals

confidence - Confidence that the statistic is within the bound, or 1-alpha, on the interval [0,1], where confidence=0 means definitely not within the bound and confidence=1 means definitely within the bound.

Returns:

ConfusionMatrixConfidenceInterval that captures the ConfidenceIntervals of the correct/incorrect classification rates