weka.associations
Class Apriori

java.lang.Object
  weka.associations.Associator
      weka.associations.Apriori

All Implemented Interfaces:

java.lang.Cloneable, OptionHandler, java.io.Serializable

public class Apriori

extends Associator

implements OptionHandler

Class implementing an Apriori-type algorithm. Iteratively reduces the minimum support until it finds the required number of rules with the given minimum confidence.

Reference: R. Agrawal, R. Srikant (1994). Fast algorithms for mining association rules in large databases . Proc International Conference on Very Large Databases, pp. 478-499. Santiage, Chile: Morgan Kaufmann, Los Altos, CA.

Valid options are:

-N required number of rules
The required number of rules (default: 10).

-T type of metric by which to sort rules
0 = confidence | 1 = lift | 2 = leverage | 3 = Conviction.

-C minimum confidence of a rule
The minimum confidence of a rule (default: 0.9).

-D delta for minimum support
The delta by which the minimum support is decreased in each iteration (default: 0.05).

-U upper bound for minimum support
The upper bound for minimum support. Don't explicitly look for rules with more than this level of support.

-M lower bound for minimum support
The lower bound for the minimum support (default = 0.1).

-S significance level
If used, rules are tested for significance at the given level. Slower (default = no significance testing).

-R
If set then columns that contain all missing values are removed from the data. -I
If set the itemsets found are also output (default = no).

Version:

$Revision: 1.15 $

Author:

Eibe Frank (eibe@cs.waikato.ac.nz), Mark Hall (mhall@cs.waikato.ac.nz)

See Also:

Serialized Form

Field Summary

protected static int	CONFIDENCE Metric types.
protected static int	CONVICTION
protected static int	LEVERAGE
protected static int	LIFT
protected FastVector[]	m_allTheRules The list of all generated rules.
protected int	m_cycles Number of cycles used before required number of rules was one.
protected double	m_delta Delta by which m_minSupport is decreased in each iteration.
protected FastVector	m_hashtables The same information stored in hash tables.
protected Instances	m_instances The instances (transactions) to be used for generating the association rules.
protected double	m_lowerBoundMinSupport The lower bound for the minimum support.
protected FastVector	m_Ls The set of all sets of itemsets L.
protected int	m_metricType The selected metric type.
protected double	m_minMetric The minimum metric score.
protected double	m_minSupport The minimum support.
protected int	m_numRules The maximum number of rules that are output.
protected boolean	m_outputItemSets Output itemsets found?
protected boolean	m_removeMissingCols
protected double	m_significanceLevel Significance level for optional significance test.
protected double	m_upperBoundMinSupport The upper bound on the support
protected boolean	m_verbose Report progress iteratively
static Tag[]	TAGS_SELECTION

Constructor Summary

Apriori()
Constructor that allows to sets default values for the minimum confidence and the maximum number of rules the minimum confidence.

Method Summary

void	buildAssociations(Instances instances) Method that generates all large itemsets with a minimum support, and from these all association rules with a minimum confidence.
java.lang.String	deltaTipText() Returns the tip text for this property
private void	findLargeItemSets(Instances instances) Method that finds all large itemsets for the given set of instances.
private void	findRulesBruteForce() Method that finds all association rules and performs significance test.
private void	findRulesQuickly() Method that finds all association rules.
double	getDelta() Get the value of delta.
double	getLowerBoundMinSupport() Get the value of lowerBoundMinSupport.
SelectedTag	getMetricType() Get the metric type
double	getMinMetric() Get the value of minConfidence.
int	getNumRules() Get the value of numRules.
java.lang.String[]	getOptions() Gets the current settings of the Apriori object.
boolean	getRemoveAllMissingCols() Returns whether columns containing all missing values are to be removed
double	getSignificanceLevel() Get the value of significanceLevel.
double	getUpperBoundMinSupport() Get the value of upperBoundMinSupport.
java.lang.String	globalInfo() Returns a string describing this associator
java.util.Enumeration	listOptions() Returns an enumeration describing the available options.
java.lang.String	lowerBoundMinSupportTipText() Returns the tip text for this property
static void	main(java.lang.String[] options) Main method for testing this class.
java.lang.String	metricTypeTipText() Returns the tip text for this property
java.lang.String	minMetricTipText() Returns the tip text for this property
java.lang.String	numRulesTipText() Returns the tip text for this property
java.lang.String	removeAllMissingColsTipText() Returns the tip text for this property
private Instances	removeMissingColumns(Instances instances) Removes columns that are all missing from the data
void	resetOptions() Resets the options to the default values.
void	setDelta(double v) Set the value of delta.
void	setLowerBoundMinSupport(double v) Set the value of lowerBoundMinSupport.
void	setMetricType(SelectedTag d) Set the metric type for ranking rules
void	setMinMetric(double v) Set the value of minConfidence.
void	setNumRules(int v) Set the value of numRules.
void	setOptions(java.lang.String[] options) Parses a given list of options.
void	setRemoveAllMissingCols(boolean r) Remove columns containing all missing values.
void	setSignificanceLevel(double v) Set the value of significanceLevel.
void	setUpperBoundMinSupport(double v) Set the value of upperBoundMinSupport.
java.lang.String	significanceLevelTipText() Returns the tip text for this property
java.lang.String	toString() Outputs the size of all the generated sets of itemsets and the rules.
java.lang.String	upperBoundMinSupportTipText() Returns the tip text for this property

Methods inherited from class weka.associations.Associator

forName, makeCopies

Methods inherited from class java.lang.Object

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

Field Detail

m_minSupport

protected double m_minSupport

The minimum support.

m_upperBoundMinSupport

protected double m_upperBoundMinSupport

The upper bound on the support

m_lowerBoundMinSupport

protected double m_lowerBoundMinSupport

The lower bound for the minimum support.

CONFIDENCE

protected static final int CONFIDENCE

Metric types.

See Also:

Constant Field Values

LIFT

protected static final int LIFT

See Also:

Constant Field Values

LEVERAGE

protected static final int LEVERAGE

See Also:

Constant Field Values

CONVICTION

protected static final int CONVICTION

See Also:

Constant Field Values

TAGS_SELECTION

public static final Tag[] TAGS_SELECTION

m_metricType

protected int m_metricType

The selected metric type.

m_minMetric

protected double m_minMetric

The minimum metric score.

m_numRules

protected int m_numRules

The maximum number of rules that are output.

m_delta

protected double m_delta

Delta by which m_minSupport is decreased in each iteration.

m_significanceLevel

protected double m_significanceLevel

Significance level for optional significance test.

m_cycles

protected int m_cycles

Number of cycles used before required number of rules was one.

m_Ls

protected FastVector m_Ls

The set of all sets of itemsets L.

m_hashtables

protected FastVector m_hashtables

The same information stored in hash tables.

m_allTheRules

protected FastVector[] m_allTheRules

The list of all generated rules.

m_instances

protected Instances m_instances

The instances (transactions) to be used for generating the association rules.

m_outputItemSets

protected boolean m_outputItemSets

Output itemsets found?

m_removeMissingCols

protected boolean m_removeMissingCols

m_verbose

protected boolean m_verbose

Report progress iteratively

Constructor Detail

Apriori

public Apriori()

Constructor that allows to sets default values for the minimum confidence and the maximum number of rules the minimum confidence.

Method Detail

globalInfo

public java.lang.String globalInfo()

Returns a string describing this associator

Returns:

a description of the evaluator suitable for displaying in the explorer/experimenter gui

resetOptions

public void resetOptions()

Resets the options to the default values.

removeMissingColumns

private Instances removeMissingColumns(Instances instances)
                                throws java.lang.Exception

Removes columns that are all missing from the data

Parameters:

instances - the instances

Returns:

a new set of instances with all missing columns removed

Throws:

java.lang.Exception

buildAssociations

public void buildAssociations(Instances instances)
                       throws java.lang.Exception

Method that generates all large itemsets with a minimum support, and from these all association rules with a minimum confidence.

Specified by:

buildAssociations in class Associator

Parameters:

instances - the instances to be used for generating the associations

Throws:

java.lang.Exception - if rules can't be built successfully

listOptions

public java.util.Enumeration listOptions()

Returns an enumeration describing the available options.

Specified by:

listOptions in interface OptionHandler

Returns:

an enumeration of all the available options.

setOptions

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

Parses a given list of options. Valid options are:

-N required number of rules
The required number of rules (default: 10).

-T type of metric by which to sort rules
0 = confidence | 1 = lift | 2 = leverage | 3 = Conviction.

-C minimum metric score of a rule
The minimum confidence of a rule (default: 0.9).

-D delta for minimum support
The delta by which the minimum support is decreased in each iteration (default: 0.05). -U upper bound for minimum support
The upper bound for minimum support. Don't explicitly look for rules with more than this level of support.

-M lower bound for minimum support
The lower bound for the minimum support (default = 0.1).

-S significance level
If used, rules are tested for significance at the given level. Slower (default = no significance testing).

-I
If set the itemsets found are also output (default = no).

-V
If set then progress is reported iteratively during execution.

-R
If set then columns that contain all missing values are removed from the data.

Specified by:

setOptions in interface OptionHandler

Parameters:

options - the list of options as an array of strings

Throws:

java.lang.Exception - if an option is not supported

getOptions

public java.lang.String[] getOptions()

Gets the current settings of the Apriori object.

Specified by:

getOptions in interface OptionHandler

Returns:

an array of strings suitable for passing to setOptions

toString

public java.lang.String toString()

Outputs the size of all the generated sets of itemsets and the rules.

removeAllMissingColsTipText

public java.lang.String removeAllMissingColsTipText()

Returns the tip text for this property

Returns:

tip text for this property suitable for displaying in the explorer/experimenter gui

setRemoveAllMissingCols

public void setRemoveAllMissingCols(boolean r)

Remove columns containing all missing values.

Parameters:

r - true if cols are to be removed.

getRemoveAllMissingCols

public boolean getRemoveAllMissingCols()

Returns whether columns containing all missing values are to be removed

Returns:

true if columns are to be removed.

upperBoundMinSupportTipText

public java.lang.String upperBoundMinSupportTipText()

Returns the tip text for this property

Returns:

tip text for this property suitable for displaying in the explorer/experimenter gui

getUpperBoundMinSupport

public double getUpperBoundMinSupport()

Get the value of upperBoundMinSupport.

Returns:

Value of upperBoundMinSupport.

setUpperBoundMinSupport

public void setUpperBoundMinSupport(double v)

Set the value of upperBoundMinSupport.

Parameters:

v - Value to assign to upperBoundMinSupport.

lowerBoundMinSupportTipText

public java.lang.String lowerBoundMinSupportTipText()

Returns the tip text for this property

Returns:

tip text for this property suitable for displaying in the explorer/experimenter gui

getLowerBoundMinSupport

public double getLowerBoundMinSupport()

Get the value of lowerBoundMinSupport.

Returns:

Value of lowerBoundMinSupport.

setLowerBoundMinSupport

public void setLowerBoundMinSupport(double v)

Set the value of lowerBoundMinSupport.

Parameters:

v - Value to assign to lowerBoundMinSupport.

getMetricType

public SelectedTag getMetricType()

Get the metric type

Returns:

the type of metric to use for ranking rules

metricTypeTipText

public java.lang.String metricTypeTipText()

Returns the tip text for this property

Returns:

tip text for this property suitable for displaying in the explorer/experimenter gui

setMetricType

public void setMetricType(SelectedTag d)

Set the metric type for ranking rules

Parameters:

d - the type of metric

minMetricTipText

public java.lang.String minMetricTipText()

Returns the tip text for this property

Returns:

tip text for this property suitable for displaying in the explorer/experimenter gui

getMinMetric

public double getMinMetric()

Get the value of minConfidence.

Returns:

Value of minConfidence.

setMinMetric

public void setMinMetric(double v)

Set the value of minConfidence.

Parameters:

v - Value to assign to minConfidence.

numRulesTipText

public java.lang.String numRulesTipText()

Returns the tip text for this property

Returns:

tip text for this property suitable for displaying in the explorer/experimenter gui

getNumRules

public int getNumRules()

Get the value of numRules.

Returns:

Value of numRules.

setNumRules

public void setNumRules(int v)

Set the value of numRules.

Parameters:

v - Value to assign to numRules.

deltaTipText

public java.lang.String deltaTipText()

Returns the tip text for this property

Returns:

tip text for this property suitable for displaying in the explorer/experimenter gui

getDelta

public double getDelta()

Get the value of delta.

Returns:

Value of delta.

setDelta

public void setDelta(double v)

Set the value of delta.

Parameters:

v - Value to assign to delta.

significanceLevelTipText

public java.lang.String significanceLevelTipText()

Returns the tip text for this property

Returns:

tip text for this property suitable for displaying in the explorer/experimenter gui

getSignificanceLevel

public double getSignificanceLevel()

Get the value of significanceLevel.

Returns:

Value of significanceLevel.

setSignificanceLevel

public void setSignificanceLevel(double v)

Set the value of significanceLevel.

Parameters:

v - Value to assign to significanceLevel.

findLargeItemSets

private void findLargeItemSets(Instances instances)
                        throws java.lang.Exception

Method that finds all large itemsets for the given set of instances.

Throws:

java.lang.Exception - if an attribute is numeric

findRulesBruteForce

private void findRulesBruteForce()
                          throws java.lang.Exception

Method that finds all association rules and performs significance test.

Throws:

java.lang.Exception - if an attribute is numeric

findRulesQuickly

private void findRulesQuickly()
                       throws java.lang.Exception

Method that finds all association rules.

Throws:

java.lang.Exception - if an attribute is numeric

main

public static void main(java.lang.String[] options)

Main method for testing this class.