com.bayesserver.learning.structure

Class FeatureSelectionTest

java.lang.Object

com.bayesserver.learning.structure.FeatureSelectionTest

public final class FeatureSelectionTest
extends Object

Contains information about a test carried out between a variable and a target to determine whether the variable is likely to be a feature or not.

Method Summary

Modifier and Type	Method and Description
Double	getMutualInformation() Gets the mutual information between the target and the test in NATS.
Double	getOneMinusPValue() Gets a value which equals one minus the p-value returned from the statical independence test.
Variable	getTarget() Gets the variable that was the target of the feature selection test.
Integer	getTemporalOrder() Gets the temporal order (if any) used to test the variables.
Variable	getVariable() Gets the variable which was tested to see if it is likely to be a feature of the getTarget() variable.
boolean	isFeature(double significanceLevel) Provides a hint as to whether the variable is likely to be a feature or not, at the given [significanceLevel].

Methods inherited from class java.lang.Object

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

Method Detail

getVariable

public Variable getVariable()

Gets the variable which was tested to see if it is likely to be a feature of the getTarget() variable.

getTemporalOrder

public Integer getTemporalOrder()

Gets the temporal order (if any) used to test the variables.

getOneMinusPValue

public Double getOneMinusPValue()

Gets a value which equals one minus the p-value returned from the statical independence test. The higher this value, the more likely the variable is to be a feature.

getMutualInformation

public Double getMutualInformation()

Gets the mutual information between the target and the test in NATS. The higher this value, the more likely the variable is to be a feature.

isFeature

public boolean isFeature(double significanceLevel)

Provides a hint as to whether the variable is likely to be a feature or not, at the given [significanceLevel].

Parameters:

significanceLevel -

Returns:

true if the variable is likely to be a feature at the given [significanceLevel], or false otherwise.

getTarget

public Variable getTarget()

Gets the variable that was the target of the feature selection test.