weka.classifiers.mi

Class MIRI

java.lang.Object

weka.classifiers.AbstractClassifier

weka.classifiers.RandomizableClassifier

weka.classifiers.mi.MITI

weka.classifiers.mi.MIRI

All Implemented Interfaces:

java.io.Serializable, java.lang.Cloneable, weka.classifiers.Classifier, weka.core.AdditionalMeasureProducer, weka.core.BatchPredictor, weka.core.CapabilitiesHandler, weka.core.CapabilitiesIgnorer, weka.core.CommandlineRunnable, weka.core.MultiInstanceCapabilitiesHandler, weka.core.OptionHandler, weka.core.Randomizable, weka.core.RevisionHandler, weka.core.TechnicalInformationHandler

public class MIRI
extends MITI
implements weka.core.OptionHandler, weka.core.AdditionalMeasureProducer

MIRI (Multi Instance Rule Inducer): multi-instance classifier that utilizes partial MITI trees witha single positive leaf to learn and represent rules. For more information, see

Hendrik Blockeel, David Page, Ashwin Srinivasan: Multi-instance Tree Learning. In: Proceedings of the International Conference on Machine Learning, 57-64, 2005.

Luke Bjerring, Eibe Frank: Beyond Trees: Adopting MITI to Learn Rules and Ensemble Classifiers for Multi-instance Data. In: Proceedings of the Australasian Joint Conference on Artificial Intelligence, 2011.

BibTeX:

 @inproceedings{Blockeel2005,
    author = {Hendrik Blockeel and David Page and Ashwin Srinivasan},
    booktitle = {Proceedings of the International Conference on Machine Learning},
    pages = {57-64},
    publisher = {ACM},
    title = {Multi-instance Tree Learning},
    year = {2005}
 }
 
 @inproceedings{Bjerring2011,
    author = {Luke Bjerring and Eibe Frank},
    booktitle = {Proceedings of the Australasian Joint Conference on Artificial Intelligence},
    publisher = {Springer},
    title = {Beyond Trees: Adopting MITI to Learn Rules and Ensemble Classifiers for Multi-instance Data},
    year = {2011}
 }
 

Valid options are:

 -M [1|2|3]
  The method used to determine best split:
  1. Gini; 2. MaxBEPP; 3. SSBEPP

 -K [kBEPPConstant]
  The constant used in the tozero() hueristic

 -L
  Scales the value of K to the size of the bags

 -U
  Use unbiased estimate rather than BEPP, i.e. UEPP.

 -B
  Uses the instances present for the bag counts at each node when splitting,
  weighted according to 1 - Ba ^ n, where n is the number of instances
  present which belong to the bag, and Ba is another parameter (default 0.5)

 -Ba [multiplier]
  Multiplier for count influence of a bag based on the number of its instances

 -A [number of attributes]
  The number of randomly selected attributes to split
  -1: All attributes
  -2: square root of the total number of attributes

 -An [number of splits]
  The number of top scoring attribute splits to randomly pick from
  -1: All splits (completely random selection)
  -2: square root of the number of splits

 -S <num>
  Random number seed.
  (default 1)

 -D
  If set, classifier is run in debug mode and
  may output additional info to the console

Author:

Luke Bjerring, Eibe Frank

See Also:

Serialized Form

Field Summary

Fields inherited from class weka.classifiers.mi.MITI

SPLITMETHOD_GINI, SPLITMETHOD_MAXBEPP, SPLITMETHOD_SSBEPP, TAGS_SPLITMETHOD

Fields inherited from class weka.classifiers.AbstractClassifier

BATCH_SIZE_DEFAULT, NUM_DECIMAL_PLACES_DEFAULT

Constructor Summary

Constructors

Constructor and Description
MIRI()

Method Summary

Modifier and Type	Method and Description
void	buildClassifier(weka.core.Instances trainingData) Generates the rule set based on the given training data.
double[]	distributionForInstance(weka.core.Instance newBag) Returns the distribution of "class probabilities" for a new bag.
java.util.Enumeration	enumerateMeasures() Returns an enumeration of the additional measure names.
double	getMeasure(java.lang.String additionalMeasureName) Returns the value of the named measure.
java.lang.String	globalInfo() Returns a string describing classifier
static void	main(java.lang.String[] options) Used to run the algorithm from the command-line.
java.lang.String	toString() Returns string representing the rule set.

Methods inherited from class weka.classifiers.mi.MITI

attributesToSplitTipText, baTipText, bTipText, getAttributesToSplit, getB, getBa, getCapabilities, getK, getL, getMultiInstanceCapabilities, getOptions, getSplitMethod, getTechnicalInformation, getTopNAttributesToSplit, getUnbiasedEstimate, kTipText, listOptions, lTipText, setAttributesToSplit, setB, setBa, setK, setL, setOptions, setSplitMethod, setTopNAttributesToSplit, setUnbiasedEstimate, splitMethodTipText, topNAttributesToSplitTipText, unbiasedEstimateTipText

Methods inherited from class weka.classifiers.RandomizableClassifier

getSeed, seedTipText, setSeed

Methods inherited from class weka.classifiers.AbstractClassifier

batchSizeTipText, classifyInstance, debugTipText, distributionsForInstances, doNotCheckCapabilitiesTipText, forName, getBatchSize, getDebug, getDoNotCheckCapabilities, getNumDecimalPlaces, getRevision, implementsMoreEfficientBatchPrediction, makeCopies, makeCopy, numDecimalPlacesTipText, postExecution, preExecution, run, runClassifier, setBatchSize, setDebug, setDoNotCheckCapabilities, setNumDecimalPlaces

Methods inherited from class java.lang.Object

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

Methods inherited from interface weka.core.OptionHandler

getOptions, listOptions, makeCopy, setOptions

Constructor Detail

MIRI

public MIRI()

Method Detail

globalInfo

public java.lang.String globalInfo()

Returns a string describing classifier

Overrides:

globalInfo in class MITI

Returns:

a description suitable for displaying in the explorer/experimenter gui

enumerateMeasures

public java.util.Enumeration enumerateMeasures()

Returns an enumeration of the additional measure names.

Specified by:

enumerateMeasures in interface weka.core.AdditionalMeasureProducer

Overrides:

enumerateMeasures in class MITI

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 weka.core.AdditionalMeasureProducer

Overrides:

getMeasure in class MITI

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

buildClassifier

public void buildClassifier(weka.core.Instances trainingData)
                     throws java.lang.Exception

Generates the rule set based on the given training data.

Specified by:

buildClassifier in interface weka.classifiers.Classifier

Overrides:

buildClassifier in class MITI

Throws:

java.lang.Exception

distributionForInstance

public double[] distributionForInstance(weka.core.Instance newBag)
                                 throws java.lang.Exception

Returns the distribution of "class probabilities" for a new bag.

Specified by:

distributionForInstance in interface weka.classifiers.Classifier

Overrides:

distributionForInstance in class MITI

Throws:

java.lang.Exception

toString

public java.lang.String toString()

Returns string representing the rule set.

Overrides:

toString in class MITI

main

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

Used to run the algorithm from the command-line.