de.tud.ke.mrapp.rulelearning.core.heuristics

Class Heuristic

java.lang.Object

de.tud.ke.mrapp.rulelearning.core.heuristics.Heuristic

All Implemented Interfaces:

java.io.Serializable

Direct Known Subclasses:

Accuracy, FMeasure, HammingLoss, JaccardIndex, Laplace, MEstimate, Precision, Recall

public abstract class Heuristic
extends java.lang.Object
implements java.io.Serializable

Defines the interface, a heuristic that calculates a heuristic value based on a ConfusionMatrix, must implement.

Since:

0.1.0

See Also:

Serialized Form

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	Heuristic.Comparator<T extends Measurable> A comparator that allows to compare Measurables based on their heuristic values according to a certain Heuristic.

Constructor Summary

Constructors

Constructor and Description
Heuristic()

Method Summary

Modifier and Type	Method and Description
boolean	equals(java.lang.Object obj)
abstract double	evaluate(@NotNull ConfusionMatrix confusionMatrix) Calculates a heuristic values based on a ConfusionMatrix.
double	evaluate(@NotNull Measurable measurable) Calculates the heuristic value of a Measurable.
int	hashCode()
abstract boolean	isGainMetric() Returns, whether the heuristic is a gain metric, i.e.
boolean	isLossMetric() Returns, whether the heuristic is a loss metric, i.e.
abstract double	maxValue() Returns the maximum heuristic value that may be calculated by the heuristic.
abstract double	minValue() Returns the minimum heuristic value that may be calculated by the heuristic.

Methods inherited from class java.lang.Object

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

Constructor Detail

Heuristic

public Heuristic()

Method Detail

evaluate

public double evaluate(@NotNull
                       @NotNull Measurable measurable)

Calculates the heuristic value of a Measurable.

Parameters:

measurable - The Measurable to be evaluated

Returns:

The heuristic value that has been calculated

evaluate

public abstract double evaluate(@NotNull
                                @NotNull ConfusionMatrix confusionMatrix)

Calculates a heuristic values based on a ConfusionMatrix.

Parameters:

confusionMatrix - The confusion matrix to be evaluated

Returns:

The heuristic value that has been calculated

isGainMetric

public abstract boolean isGainMetric()

Returns, whether the heuristic is a gain metric, i.e. that greater heuristic values are considered to be better than lower heuristic values, or not.

Returns:

True, if the heuristic is a gain metric, false otherwise

isLossMetric

public boolean isLossMetric()

Returns, whether the heuristic is a loss metric, i.e. that lower heuristic values are considered to be better than greater heuristic values, or not.

Returns:

True, if the heuristic is a loss metric, false otherwise

maxValue

public abstract double maxValue()

Returns the maximum heuristic value that may be calculated by the heuristic.

Returns:

The maximum heuristic value that may be calculated by the heuristic

minValue

public abstract double minValue()

Returns the minimum heuristic value that may be calculated by the heuristic.

Returns:

The minimum heuristic value that may be calculated by the heuristic

hashCode

public int hashCode()

Overrides:

hashCode in class java.lang.Object

equals

public boolean equals(java.lang.Object obj)

Overrides:

equals in class java.lang.Object