Package adaa.analytics.rules.logic.induction

Class RegressionFinder

java.lang.Object

adaa.analytics.rules.logic.induction.AbstractFinder

adaa.analytics.rules.logic.induction.RegressionFinder

All Implemented Interfaces:

AutoCloseable

Direct Known Subclasses:

ContrastRegressionFinder, RegressionExpertFinder, SurvivalLogRankFinder

public class RegressionFinder
extends AbstractFinder

Algorithm for growing and pruning regression rules.

Field Summary

Fields inherited from class adaa.analytics.rules.logic.induction.AbstractFinder

attributeValuesOrder, modifier, params, pool, threadCount

Constructor Summary

Constructors

Constructor	Description
RegressionFinder(InductionParameters params)

Method Summary

Modifier and Type	Method	Description
protected boolean	checkCandidate(IExampleSet dataset, Rule rule, ConditionBase candidate, Set<Integer> uncovered, Set<Integer> covered, ConditionEvaluation currentBest)
protected boolean	checkCandidate(ElementaryCondition cnd, double p, double n, double new_p, double new_n, double P, double N, double uncoveredSize, int ruleOrderNum)
protected ElementaryCondition	induceCondition(Rule rule, IExampleSet dataset, Set<Integer> uncovered, Set<Integer> covered, Set<IAttribute> allowedAttributes, Pair<String,Object>... extraParams)	Abstract method representing all procedures which induce an elementary condition.
protected ElementaryCondition	induceCondition_mean(Rule rule, IExampleSet dataset, Set<Integer> uncovered, Set<Integer> covered, Set<IAttribute> allowedAttributes, Pair<String,Object>... extraParams)

Methods inherited from class adaa.analytics.rules.logic.induction.AbstractFinder

addObserver, clearObservers, close, grow, names2attributes, notifyConditionAdded, notifyConditionRemoved, notifyGrowingFinished, notifyGrowingStarted, notifyRuleReady, postprocess, preprocess, prune

Methods inherited from class java.lang.Object

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

Constructor Details

RegressionFinder

public RegressionFinder(InductionParameters params)

Method Details

induceCondition_mean

protected ElementaryCondition induceCondition_mean(Rule rule,
 IExampleSet dataset,
 Set<Integer> uncovered,
 Set<Integer> covered,
 Set<IAttribute> allowedAttributes,
 Pair<String,Object>... extraParams)

induceCondition

protected ElementaryCondition induceCondition(Rule rule,
 IExampleSet dataset,
 Set<Integer> uncovered,
 Set<Integer> covered,
 Set<IAttribute> allowedAttributes,
 Pair<String,Object>... extraParams)

Description copied from class: AbstractFinder

Abstract method representing all procedures which induce an elementary condition.

Specified by:

induceCondition in class AbstractFinder

Parameters:

rule - Current rule.

dataset - Training set.

uncovered - Set of examples uncovered by the model.

covered - Set of examples covered by the rule being grown.

allowedAttributes - Set of attributes that may be used during induction.

extraParams - Additional parameters.

Returns:

Induced elementary condition.

checkCandidate

protected boolean checkCandidate(IExampleSet dataset,
 Rule rule,
 ConditionBase candidate,
 Set<Integer> uncovered,
 Set<Integer> covered,
 ConditionEvaluation currentBest)

checkCandidate

protected boolean checkCandidate(ElementaryCondition cnd,
 double p,
 double n,
 double new_p,
 double new_n,
 double P,
 double N,
 double uncoveredSize,
 int ruleOrderNum)