weka.classifiers.rules
Class JRip

java.lang.Object
  extended byweka.classifiers.Classifier
      extended byweka.classifiers.DistributionClassifier
          extended byweka.classifiers.rules.JRip
All Implemented Interfaces:
AdditionalMeasureProducer, java.lang.Cloneable, OptionHandler, java.io.Serializable, WeightedInstancesHandler

public class JRip
extends DistributionClassifier
implements OptionHandler, AdditionalMeasureProducer, WeightedInstancesHandler

This class implements a propositional rule learner, Repeated Incremental Pruning to Produce Error Reduction (RIPPER), which is proposed by William W. Cohen as an optimzed version of IREP.

The algorithm is briefly described as follows:

Initialize RS = {}, and for each class from the less prevalent one to the more frequent one, DO:

1. Building stage: repeat 1.1 and 1.2 until the descrition length (DL) of the ruleset and examples is 64 bits greater than the smallest DL met so far, or there are no positive examples, or the error rate >= 50%.

1.1. Grow phase:
Grow one rule by greedily adding antecedents (or conditions) to the rule until the rule is perfect (i.e. 100% accurate). The procedure tries every possible value of each attribute and selects the condition with highest information gain: p(log(p/t)-log(P/T)).

1.2. Prune phase:
Incrementally prune each rule and allow the pruning of any final sequences of the antecedents;
The pruning metric is (p-n)/(p+n) -- but it's actually 2p/(p+n) -1, so in this implementation we simply use p/(p+n) (actually (p+1)/(p+n+2), thus if p+n is 0, it's 0.5).

2. Optimization stage: after generating the initial ruleset {Ri}, generate and prune two variants of each rule Ri from randomized data using procedure 1.1 and 1.2. But one variant is generated from an empty rule while the other is generated by greedily adding antecedents to the original rule. Moreover, the pruning metric used here is (TP+TN)/(P+N).
Then the smallest possible DL for each variant and the original rule is computed. The variant with the minimal DL is selected as the final representative of Ri in the ruleset.
After all the rules in {Ri} have been examined and if there are still residual positives, more rules are generated based on the residual positives using Building Stage again.

3. Delete the rules from the ruleset that would increase the DL of the whole ruleset if it were in it. and add resultant ruleset to RS.

ENDDO

Note that there seem to be 2 bugs in the ripper program that would affect the ruleset size and accuracy slightly. This implementation avoids these bugs and thus is a little bit different from Cohen's original implementation. Even after fixing the bugs, since the order of classes with the same frequency is not defined in ripper, there still seems to be some trivial difference between this implementation and the original ripper, especially for audiology data in UCI repository, where there are lots of classes of few instances.

If wrapped by other classes, typical usage of this class is:
JRip rip = new JRip(); Instances data = ... // Data from somewhere double[] orderedClasses = ... // Get the ordered class counts for the data double expFPRate = ... // Calculate the expected FP/(FP+FN) rate double classIndex = ... // The class index for which ruleset is built // DL of default rule, no theory DL, only data DL double defDL = RuleStats.dataDL(expFPRate, 0.0, data.sumOfWeights(), 0.0, orderedClasses[(int)classIndex]); rip.rulesetForOneClass(expFPRate, data, classIndex, defDL); RuleStats rulesetStats = rip.getRuleStats(0); // Can get heaps of information from RuleStats, e.g. combined DL, // simpleStats, etc. double comDL = rulesetStats.combinedDL(expFPRate, classIndex); int whichRule = ... // Want simple stats of which rule? double[] simpleStats = rulesetStats.getSimpleStats(whichRule); ... Details please see "Fast Effective Rule Induction", William W. Cohen, 'Machine Learning: Proceedings of the Twelfth International Conference' (ML95).

PS. We have compared this implementation with the original ripper implementation in aspects of accuracy, ruleset size and running time on both artificial data "ab+bcd+defg" and UCI datasets. In all these aspects it seems to be quite comparable to the original ripper implementation. However, we didn't consider memory consumption optimization in this implementation.

See Also:
Serialized Form

Nested Class Summary
protected  class JRip.RipperRule
          This class implements a single rule that predicts specified class.
 
Constructor Summary
JRip()
           
 
Method Summary
 void buildClassifier(Instances instances)
          Builds Ripper in the order of class frequencies.
 double[] distributionForInstance(Instance datum)
          Classify the test instance with the rule learner and provide the class distributions
 java.util.Enumeration enumerateMeasures()
          Returns an enumeration of the additional measure names
 boolean getCheckErrorRate()
           
 boolean getDebug()
           
 int getFolds()
           
 double getMeasure(java.lang.String additionalMeasureName)
          Returns the value of the named measure
 double getMinNo()
           
 int getOptimizations()
           
 java.lang.String[] getOptions()
          Gets the current settings of the Classifier.
 FastVector getRuleset()
          Get the ruleset generated by Ripper
 RuleStats getRuleStats(int pos)
          Get the statistics of the ruleset in the given position
 long getSeed()
           
 boolean getUsePruning()
           
 java.util.Enumeration listOptions()
          Returns an enumeration describing the available options Valid options are:
static void main(java.lang.String[] args)
          Main method.
protected  Instances rulesetForOneClass(double expFPRate, Instances data, double classIndex, double defDL)
          Build a ruleset for the given class according to the given data
 void setCheckErrorRate(boolean d)
           
 void setDebug(boolean d)
           
 void setFolds(int fold)
           
 void setMinNo(double m)
           
 void setOptimizations(int run)
           
 void setOptions(java.lang.String[] options)
          Parses a given list of options.
 void setSeed(long s)
           
 void setUsePruning(boolean d)
           
 java.lang.String toString()
          Prints the all the rules of the rule learner.
 
Methods inherited from class weka.classifiers.DistributionClassifier
calculateEntropy, calculateLabeledInstanceMargin, calculateMargin, classifyInstance
 
Methods inherited from class weka.classifiers.Classifier
forName, makeCopies
 
Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait
 

Constructor Detail

JRip

public JRip()
Method Detail

listOptions

public java.util.Enumeration listOptions()
Returns an enumeration describing the available options Valid options are:

-F number
The number of folds for reduced error pruning. One fold is used as the pruning set. (Default: 3)

-N number
The minimal weights of instances within a split. (Default: 2)

-O number
Set the number of runs of optimizations. (Default: 2)

-D
Whether turn on the debug mode -S number
The seed of randomization used in Ripper.(Default: 1)

-E
Whether NOT check the error rate >= 0.5 in stopping criteria. (default: check)

-P
Whether NOT use pruning. (default: use pruning)

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.

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 Classifier.

Specified by:
getOptions in interface OptionHandler
Returns:
an array of strings suitable for passing to setOptions

enumerateMeasures

public java.util.Enumeration enumerateMeasures()
Returns an enumeration of the additional measure names

Specified by:
enumerateMeasures in interface AdditionalMeasureProducer
Returns:
an enumeration of the measure names

getMeasure

public double getMeasure(java.lang.String additionalMeasureName)
Returns the value of the named measure

Specified by:
getMeasure in interface AdditionalMeasureProducer
Parameters:
additionalMeasureName - the name of the measure to query for its value
Returns:
the value of the named measure
Throws:
java.lang.IllegalArgumentException - if the named measure is not supported

setFolds

public void setFolds(int fold)

getFolds

public int getFolds()

setMinNo

public void setMinNo(double m)

getMinNo

public double getMinNo()

setSeed

public void setSeed(long s)

getSeed

public long getSeed()

setOptimizations

public void setOptimizations(int run)

getOptimizations

public int getOptimizations()

setDebug

public void setDebug(boolean d)

getDebug

public boolean getDebug()

setCheckErrorRate

public void setCheckErrorRate(boolean d)

getCheckErrorRate

public boolean getCheckErrorRate()

setUsePruning

public void setUsePruning(boolean d)

getUsePruning

public boolean getUsePruning()

getRuleset

public FastVector getRuleset()
Get the ruleset generated by Ripper

Returns:
the ruleset

getRuleStats

public RuleStats getRuleStats(int pos)
Get the statistics of the ruleset in the given position

Parameters:
pos - the position of the stats, assuming correct

buildClassifier

public void buildClassifier(Instances instances)
                     throws java.lang.Exception
Builds Ripper in the order of class frequencies. For each class it's built in two stages: building and optimization

Specified by:
buildClassifier in class Classifier
Parameters:
instances - the training data
Throws:
java.lang.Exception - if classifier can't be built successfully

distributionForInstance

public double[] distributionForInstance(Instance datum)
Classify the test instance with the rule learner and provide the class distributions

Specified by:
distributionForInstance in class DistributionClassifier
Parameters:
datum - the instance to be classified
Returns:
the distribution

rulesetForOneClass

protected Instances rulesetForOneClass(double expFPRate,
                                       Instances data,
                                       double classIndex,
                                       double defDL)
                                throws java.lang.Exception
Build a ruleset for the given class according to the given data

Parameters:
expFPRate - the expected FP/(FP+FN) used in DL calculation
data - the given data
classIndex - the given class index
defDL - the default DL in the data
Throws:
if - the ruleset can be built properly
java.lang.Exception

toString

public java.lang.String toString()
Prints the all the rules of the rule learner.

Returns:
a textual description of the classifier

main

public static void main(java.lang.String[] args)
Main method.

Parameters:
args - the options for the classifier