edu.udo.cs.yale.operator.learner.igss.utility
Class Linear

java.lang.Object
  edu.udo.cs.yale.operator.learner.igss.utility.AbstractUtility
      edu.udo.cs.yale.operator.learner.igss.utility.Linear

All Implemented Interfaces:

Utility

public class Linear
extends AbstractUtility

The utility function Linear.

Version:

$Id: Linear.java,v 1.1 2006/10/02 16:07:49 ingomierswa Exp $

Author:

Dirk Dach

Field Summary

Fields inherited from class edu.udo.cs.yale.operator.learner.igss.utility.AbstractUtility

large, priors

Fields inherited from interface edu.udo.cs.yale.operator.learner.igss.utility.Utility

FIRST_TYPE_INDEX, LAST_TYPE_INDEX, TYPE_ACCURACY, TYPE_BINOMIAL, TYPE_LINEAR, TYPE_SQUARED, TYPE_WRACC, UTILITY_TYPES

Constructor Summary

Linear(double[] priors, int large)
Constructs a new Linear with the given default probability.

Method Summary

double	conf(double totalExampleWeight, double delta) Calculate confidence intervall without a specific rule.
double	conf(double totalExampleWeight, double totalPositiveWeight, Hypothesis hypo, double delta) Calculate confidence intervall for a specific rule.
double	confSmallM(double totalExampleWeight, double delta) Calculate confidence intervall without a specific rule for small m.
double	getUpperBound(double totalWeight, double totalPositiveWeight, Hypothesis hypo, double delta) Returns an upper bound for the utility of refinements for the given hypothesis.
double	utility(double totalWeight, double totalPositiveWeight, Hypothesis hypo) Calculates the utility for the given number of examples,positive examples and hypothesis.
private double	variance(double p, double totalExampleWeight) Calculates the variance for a binomial distribution.

Methods inherited from class edu.udo.cs.yale.operator.learner.igss.utility.AbstractUtility

calculateM, confidenceIntervall, confidenceIntervall, inverseNormal

Methods inherited from class java.lang.Object

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

Constructor Detail

Linear

public Linear(double[] priors,
              int large)

Constructs a new Linear with the given default probability.

Method Detail

utility

public double utility(double totalWeight,
                      double totalPositiveWeight,
                      Hypothesis hypo)

Calculates the utility for the given number of examples,positive examples and hypothesis.

conf

public double conf(double totalExampleWeight,
                   double delta)

Calculate confidence intervall without a specific rule.

Specified by:

conf in class AbstractUtility

conf

public double conf(double totalExampleWeight,
                   double totalPositiveWeight,
                   Hypothesis hypo,
                   double delta)

Calculate confidence intervall for a specific rule.

Specified by:

conf in class AbstractUtility

variance

private double variance(double p,
                        double totalExampleWeight)

Calculates the variance for a binomial distribution.

confSmallM

public double confSmallM(double totalExampleWeight,
                         double delta)

Calculate confidence intervall without a specific rule for small m.

Specified by:

confSmallM in class AbstractUtility

getUpperBound

public double getUpperBound(double totalWeight,
                            double totalPositiveWeight,
                            Hypothesis hypo,
                            double delta)

Returns an upper bound for the utility of refinements for the given hypothesis.