Package adaa.analytics.rules.logic.performance

Class CorrelationCriterion

java.lang.Object

adaa.analytics.rules.logic.performance.AbstractPerformanceCounter

adaa.analytics.rules.logic.performance.CorrelationCriterion

public class CorrelationCriterion
extends AbstractPerformanceCounter

Computes the empirical corelation coefficient 'r' between label and prediction. For P=prediction, L=label, V=Variance, Cov=Covariance we calculate r by:
Cov(L,P) / sqrt(V(L)*V(P)).

Implementation hint: this implementation intensionally recomputes the mean and variance of prediction and label despite the fact that they are available by the Attribute objects. The reason: it can happen, that there are some examples which have a NaN as prediction or label, but not both. In this case, mean and variance stored in tie Attributes and computed here can differ.

Constructor Summary

Constructors

Constructor	Description
CorrelationCriterion()

Method Summary

Modifier and Type	Method	Description
PerformanceResult	countExample(IExampleSet eset)	Updates all sums needed to compute the correlation coefficient.

Methods inherited from class java.lang.Object

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

Constructor Details

CorrelationCriterion

public CorrelationCriterion()

Method Details

countExample

public PerformanceResult countExample(IExampleSet eset)

Updates all sums needed to compute the correlation coefficient.

Specified by:

countExample in class AbstractPerformanceCounter