weka.associations.classification

Class PrecedencePruning

java.lang.Object

weka.associations.classification.PruneCAR

weka.associations.classification.CrTree

weka.associations.classification.JCBAPruning

weka.associations.classification.PrecedencePruning

All Implemented Interfaces:

java.io.Serializable, weka.core.OptionHandler

public class PrecedencePruning
extends JCBAPruning
implements weka.core.OptionHandler, java.io.Serializable

Class that allows a simple pessimistic-error-rate based pruning or omits pruning at all and just builds a CrTree with all mined rules. In addition to the CrTree a list containing information about the rule sorting is maintained (precedence list). This list allows investigating the sort oder of rules induced by the mining algorithm even when no pruning is performed. The Tree Structure is described in: W. Li, J. Han, J.Pei: CMAR: Accurate and Efficient Classification Based on Multiple Class-Association Rules. In ICDM'01:369-376,2001. The optional pessimistic-error-rate-based pruning step is described in: B. Liu, W. Hsu, Y. Ma: Integrating Classification and Association Rule Mining. In KDD'98:80-86,1998. Valid options are:

-C the confidence value
The confidence value for the optional pessimistic-error-rate-based pruning step (default: 0.25).

-N
If set no pessimistic-error-rate-based pruning is performed.

Version:

$Revision: 10209 $

Author:

Stefan Mutter

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
PrecedencePruning() Constructor

Method Summary

Modifier and Type	Method and Description
int	getDefaultClass() Gets the default class
java.lang.String[]	getOptions() Gets options
void	insertContent(CrNode node, weka.core.FastVector input) Insert consequence to a node and updates the precedence list
java.util.Enumeration	listOptions() Lists options
int	numClassRules() Gets the number of classification rules
int	numMinedRules() Gets the number of mined rules
int	numPrunedRules() Gets the number of rules after the optional pessimistic-error.rate-based pruning
void	preprocess(java.util.ArrayList<java.lang.Object> premises, java.util.ArrayList<java.lang.Object> consequences, java.util.ArrayList<java.lang.Object> confidences) Preprocesses a rule before it gets inserted into the tree.
void	prune() No pruning is perfomed (except of the optional pruning step in the preprocessing method).
void	setDefaultClass() Sets the default class as the majority class label of all instances
void	setNoPessPruning(boolean flag) Sets pessimistic-error-rate-based pruning
void	setOptions(java.lang.String[] options) Sets options
java.lang.String	toString(java.lang.String metricString, int numMinedRules) Outputs the CrTree

Methods inherited from class weka.associations.classification.JCBAPruning

addErrs, calculateDefaultClass, calculateError, getClassValue, getPrecedenceList, getStopIndex, optPruning, pruneBeforeInsertion, resetOptions, setDefaultClass, setInstances, toString

Methods inherited from class weka.associations.classification.CrTree

deleteContent, deleteNode, getAssociateList, getRoot, insertNode, isEmpty, makeEmpty, pathToString, prunedRules, pruningCriterions, removeAtChild, removeAtList, removeAtSibling, reportSubtreeCount, rulePremise, search, setInstancesNoClass, setInstancesOnlyClass, sortItemSet, updateHeight

Methods inherited from class weka.associations.classification.PruneCAR

forName

Methods inherited from class java.lang.Object

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

Methods inherited from interface weka.core.OptionHandler

makeCopy

Constructor Detail

PrecedencePruning

public PrecedencePruning()

Constructor

Method Detail

insertContent

public void insertContent(CrNode node,
                          weka.core.FastVector input)

Insert consequence to a node and updates the precedence list

Overrides:

insertContent in class JCBAPruning

Parameters:

node - the node

input - the consequence and additional measures

preprocess

public void preprocess(java.util.ArrayList<java.lang.Object> premises,
                       java.util.ArrayList<java.lang.Object> consequences,
                       java.util.ArrayList<java.lang.Object> confidences)
                throws java.lang.Exception

Preprocesses a rule before it gets inserted into the tree. If set, pessimistic-error-rate based pruning is performed.

Overrides:

preprocess in class JCBAPruning

Parameters:

premises - the premises

consequences - the consequences

confidences - the metrices

Throws:

java.lang.Exception - throws exception is preprocessing is not possible.

prune

public void prune()

No pruning is perfomed (except of the optional pruning step in the preprocessing method).

Overrides:

prune in class JCBAPruning

setDefaultClass

public void setDefaultClass()

Sets the default class as the majority class label of all instances

Overrides:

setDefaultClass in class CrTree

getDefaultClass

public int getDefaultClass()

Gets the default class

Overrides:

getDefaultClass in class CrTree

Returns:

the default class

getOptions

public java.lang.String[] getOptions()

Gets options

Specified by:

getOptions in interface weka.core.OptionHandler

Overrides:

getOptions in class JCBAPruning

Returns:

the options

listOptions

public java.util.Enumeration listOptions()

Lists options

Specified by:

listOptions in interface weka.core.OptionHandler

Overrides:

listOptions in class JCBAPruning

Returns:

list of all available options

setNoPessPruning

public void setNoPessPruning(boolean flag)

Sets pessimistic-error-rate-based pruning

Parameters:

flag - true, if no pessimistic-error-rate-based pruning should be done, false otherwise

setOptions

public void setOptions(java.lang.String[] options)
                throws java.lang.Exception

Sets options

Specified by:

setOptions in interface weka.core.OptionHandler

Overrides:

setOptions in class JCBAPruning

Parameters:

options - the options

Throws:

java.lang.Exception - exception if options cannot be set.

toString

public java.lang.String toString(java.lang.String metricString,
                                 int numMinedRules)

Outputs the CrTree

Parameters:

metricString - the metric used to mine class association rules

Returns:

the CrTree as a a string

numMinedRules

public int numMinedRules()

Gets the number of mined rules

Overrides:

numMinedRules in class JCBAPruning

Returns:

the number of mined rules

numPrunedRules

public int numPrunedRules()

Gets the number of rules after the optional pessimistic-error.rate-based pruning

Overrides:

numPrunedRules in class JCBAPruning

Returns:

the number of rules after the optional pessimistic-error.rate-based pruning

numClassRules

public int numClassRules()

Gets the number of classification rules

Overrides:

numClassRules in class JCBAPruning

Returns:

the number of classification rules