edu.umass.cs.mallet.base.classify.evaluate

Class ConfusionMatrix

java.lang.Object

edu.umass.cs.mallet.base.classify.evaluate.ConfusionMatrix

public class ConfusionMatrix
extends Object

Calculates and prints confusion matrix, accuracy, and precision for a given clasification trial.

Constructor Summary

Constructors

Constructor and Description
ConfusionMatrix(Trial t) Constructs matrix and calculates values

Method Summary

Modifier and Type	Method and Description
double	getClassPrior(int classIndex) Returns the percentage of instances with true label = classIndex
double	getConfusionBetween(int class1, int class2) Returns percent of time that class2 is true class when class1 is predicted class
double	getPrecision(int predictedClassIndex) Returns the precision of this predicted class
String	toString()

Methods inherited from class java.lang.Object

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

Constructor Detail

ConfusionMatrix

public ConfusionMatrix(Trial t)

Constructs matrix and calculates values

Parameters:

t - the trial to build matrix from

Method Detail

toString

public String toString()

Overrides:

toString in class Object

getPrecision

public double getPrecision(int predictedClassIndex)

Returns the precision of this predicted class

getConfusionBetween

public double getConfusionBetween(int class1,
                                  int class2)

Returns percent of time that class2 is true class when class1 is predicted class

getClassPrior

public double getClassPrior(int classIndex)

Returns the percentage of instances with true label = classIndex