Class Decorate

  • All Implemented Interfaces:
    java.io.Serializable, java.lang.Cloneable, CapabilitiesHandler, OptionHandler, Randomizable, RevisionHandler, TechnicalInformationHandler

    public class Decorate
    extends RandomizableIteratedSingleClassifierEnhancer
    implements TechnicalInformationHandler
    DECORATE is a meta-learner for building diverse ensembles of classifiers by using specially constructed artificial training examples. Comprehensive experiments have demonstrated that this technique is consistently more accurate than the base classifier, Bagging and Random Forests.Decorate also obtains higher accuracy than Boosting on small training sets, and achieves comparable performance on larger training sets.

    For more details see:

    P. Melville, R. J. Mooney: Constructing Diverse Classifier Ensembles Using Artificial Training Examples. In: Eighteenth International Joint Conference on Artificial Intelligence, 505-510, 2003.

    P. Melville, R. J. Mooney (2004). Creating Diversity in Ensembles Using Artificial Data. Information Fusion: Special Issue on Diversity in Multiclassifier Systems..

    BibTeX:

     @inproceedings{Melville2003,
        author = {P. Melville and R. J. Mooney},
        booktitle = {Eighteenth International Joint Conference on Artificial Intelligence},
        pages = {505-510},
        title = {Constructing Diverse Classifier Ensembles Using Artificial Training Examples},
        year = {2003}
     }
     
     @article{Melville2004,
        author = {P. Melville and R. J. Mooney},
        journal = {Information Fusion: Special Issue on Diversity in Multiclassifier Systems},
        note = {submitted},
        title = {Creating Diversity in Ensembles Using Artificial Data},
        year = {2004}
     }
     

    Valid options are:

     -E
      Desired size of ensemble.
      (default 15)
     -R
      Factor that determines number of artificial examples to generate.
      Specified proportional to training set size.
      (default 1.0)
     -S <num>
      Random number seed.
      (default 1)
     -I <num>
      Number of iterations.
      (default 50)
     -D
      If set, classifier is run in debug mode and
      may output additional info to the console
     -W
      Full name of base classifier.
      (default: weka.classifiers.trees.J48)
     
     Options specific to classifier weka.classifiers.trees.J48:
     
     -U
      Use unpruned tree.
     -C <pruning confidence>
      Set confidence threshold for pruning.
      (default 0.25)
     -M <minimum number of instances>
      Set minimum number of instances per leaf.
      (default 2)
     -R
      Use reduced error pruning.
     -N <number of folds>
      Set number of folds for reduced error
      pruning. One fold is used as pruning set.
      (default 3)
     -B
      Use binary splits only.
     -S
      Don't perform subtree raising.
     -L
      Do not clean up after the tree has been built.
     -A
      Laplace smoothing for predicted probabilities.
     -Q <seed>
      Seed for random data shuffling (default 1).
    Options after -- are passed to the designated classifier.

    Version:
    $Revision: 8037 $
    Author:
    Prem Melville (melville@cs.utexas.edu)
    See Also:
    Serialized Form
    • Constructor Detail

      • Decorate

        public Decorate()
        Constructor.
    • Method Detail

      • setOptions

        public void setOptions​(java.lang.String[] options)
                        throws java.lang.Exception
        Parses a given list of options.

        Valid options are:

         -E
          Desired size of ensemble.
          (default 15)
         -R
          Factor that determines number of artificial examples to generate.
          Specified proportional to training set size.
          (default 1.0)
         -S <num>
          Random number seed.
          (default 1)
         -I <num>
          Number of iterations.
          (default 50)
         -D
          If set, classifier is run in debug mode and
          may output additional info to the console
         -W
          Full name of base classifier.
          (default: weka.classifiers.trees.J48)
         
         Options specific to classifier weka.classifiers.trees.J48:
         
         -U
          Use unpruned tree.
         -C <pruning confidence>
          Set confidence threshold for pruning.
          (default 0.25)
         -M <minimum number of instances>
          Set minimum number of instances per leaf.
          (default 2)
         -R
          Use reduced error pruning.
         -N <number of folds>
          Set number of folds for reduced error
          pruning. One fold is used as pruning set.
          (default 3)
         -B
          Use binary splits only.
         -S
          Don't perform subtree raising.
         -L
          Do not clean up after the tree has been built.
         -A
          Laplace smoothing for predicted probabilities.
         -Q <seed>
          Seed for random data shuffling (default 1).
        Options after -- are passed to the designated classifier.

        Specified by:
        setOptions in interface OptionHandler
        Overrides:
        setOptions in class RandomizableIteratedSingleClassifierEnhancer
        Parameters:
        options - the list of options as an array of strings
        Throws:
        java.lang.Exception - if an option is not supported
      • desiredSizeTipText

        public java.lang.String desiredSizeTipText()
        Returns the tip text for this property
        Returns:
        tip text for this property suitable for displaying in the explorer/experimenter gui
      • numIterationsTipText

        public java.lang.String numIterationsTipText()
        Returns the tip text for this property
        Overrides:
        numIterationsTipText in class IteratedSingleClassifierEnhancer
        Returns:
        tip text for this property suitable for displaying in the explorer/experimenter gui
      • artificialSizeTipText

        public java.lang.String artificialSizeTipText()
        Returns the tip text for this property
        Returns:
        tip text for this property suitable for displaying in the explorer/experimenter gui
      • globalInfo

        public java.lang.String globalInfo()
        Returns a string describing classifier
        Returns:
        a description suitable for displaying in the explorer/experimenter gui
      • getTechnicalInformation

        public TechnicalInformation getTechnicalInformation()
        Returns an instance of a TechnicalInformation object, containing detailed information about the technical background of this class, e.g., paper reference or book this class is based on.
        Specified by:
        getTechnicalInformation in interface TechnicalInformationHandler
        Returns:
        the technical information about this class
      • getArtificialSize

        public double getArtificialSize()
        Factor that determines number of artificial examples to generate.
        Returns:
        factor that determines number of artificial examples to generate
      • setArtificialSize

        public void setArtificialSize​(double newArtSize)
        Sets factor that determines number of artificial examples to generate.
        Parameters:
        newArtSize - factor that determines number of artificial examples to generate
      • getDesiredSize

        public int getDesiredSize()
        Gets the desired size of the committee.
        Returns:
        the desired size of the committee
      • setDesiredSize

        public void setDesiredSize​(int newDesiredSize)
        Sets the desired size of the committee.
        Parameters:
        newDesiredSize - the desired size of the committee
      • buildClassifier

        public void buildClassifier​(Instances data)
                             throws java.lang.Exception
        Build Decorate classifier
        Overrides:
        buildClassifier in class IteratedSingleClassifierEnhancer
        Parameters:
        data - the training data to be used for generating the classifier
        Throws:
        java.lang.Exception - if the classifier could not be built successfully
      • distributionForInstance

        public double[] distributionForInstance​(Instance instance)
                                         throws java.lang.Exception
        Calculates the class membership probabilities for the given test instance.
        Overrides:
        distributionForInstance in class Classifier
        Parameters:
        instance - the instance to be classified
        Returns:
        predicted class probability distribution
        Throws:
        java.lang.Exception - if distribution can't be computed successfully
      • toString

        public java.lang.String toString()
        Returns description of the Decorate classifier.
        Overrides:
        toString in class java.lang.Object
        Returns:
        description of the Decorate classifier as a string
      • main

        public static void main​(java.lang.String[] argv)
        Main method for testing this class.
        Parameters:
        argv - the options