weka.filters.unsupervised.attribute
Class RandomProjection

java.lang.Object
  extended by weka.filters.Filter
      extended by weka.filters.unsupervised.attribute.RandomProjection
All Implemented Interfaces:
java.io.Serializable, CapabilitiesHandler, OptionHandler, RevisionHandler, TechnicalInformationHandler, UnsupervisedFilter

public class RandomProjection
extends Filter
implements UnsupervisedFilter, OptionHandler, TechnicalInformationHandler

Reduces the dimensionality of the data by projecting it onto a lower dimensional subspace using a random matrix with columns of unit length (i.e. It will reduce the number of attributes in the data while preserving much of its variation like PCA, but at a much less computational cost).
It first applies the NominalToBinary filter to convert all attributes to numeric before reducing the dimension. It preserves the class attribute.

For more information, see:

Dmitriy Fradkin, David Madigan: Experiments with random projections for machine learning. In: KDD '03: Proceedings of the ninth ACM SIGKDD international conference on Knowledge discovery and data mining, New York, NY, USA, 517-522, 003.

BibTeX:

 @inproceedings{Fradkin003,
    address = {New York, NY, USA},
    author = {Dmitriy Fradkin and David Madigan},
    booktitle = {KDD '03: Proceedings of the ninth ACM SIGKDD international conference on Knowledge discovery and data mining},
    pages = {517-522},
    publisher = {ACM Press},
    title = {Experiments with random projections for machine learning},
    year = {003}
 }
 

Valid options are:

 -N <number>
  The number of dimensions (attributes) the data should be reduced to
  (default 10; exclusive of the class attribute, if it is set).
 -D [SPARSE1|SPARSE2|GAUSSIAN]
  The distribution to use for calculating the random matrix.
  Sparse1 is:
    sqrt(3)*{-1 with prob(1/6), 0 with prob(2/3), +1 with prob(1/6)}
  Sparse2 is:
    {-1 with prob(1/2), +1 with prob(1/2)}
 
 -P <percent>
  The percentage of dimensions (attributes) the data should
  be reduced to (exclusive of the class attribute, if it is set). This -N
  option is ignored if this option is present or is greater
  than zero.
 -M
  Replace missing values using the ReplaceMissingValues filter
 -R <num>
  The random seed for the random number generator used for
  calculating the random matrix (default 42).

Version:
$Revision: 1.11 $ [1.0 - 22 July 2003 - Initial version (Ashraf M. Kibriya)]
Author:
Ashraf M. Kibriya (amk14@cs.waikato.ac.nz)
See Also:
Serialized Form

Field Summary
static int GAUSSIAN
          distribution type: gaussian
static int SPARSE1
          distribution type: sparse 1
static int SPARSE2
          distribution type: sparse 2
static Tag[] TAGS_DSTRS_TYPE
          The types of distributions that can be used for calculating the random matrix
 
Constructor Summary
RandomProjection()
           
 
Method Summary
 boolean batchFinished()
          Signify that this batch of input to the filter is finished.
 java.lang.String distributionTipText()
          Returns the tip text for this property
 Capabilities getCapabilities()
          Returns the Capabilities of this filter.
 SelectedTag getDistribution()
          Returns the current distribution that'll be used for calculating the random matrix
 int getNumberOfAttributes()
          Gets the current number of attributes (dimensionality) to which the data will be reduced to.
 java.lang.String[] getOptions()
          Gets the current settings of the filter.
 double getPercent()
          Gets the percent the attributes (dimensions) of the data will be reduced to
 long getRandomSeed()
          Gets the random seed of the random number generator
 boolean getReplaceMissingValues()
          Gets the current setting for using ReplaceMissingValues filter
 java.lang.String getRevision()
          Returns the revision string.
 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.
 java.lang.String globalInfo()
          Returns a string describing this filter
 boolean input(Instance instance)
          Input an instance for filtering.
 java.util.Enumeration listOptions()
          Returns an enumeration describing the available options.
static void main(java.lang.String[] argv)
          Main method for testing this class.
 java.lang.String numberOfAttributesTipText()
          Returns the tip text for this property
 java.lang.String percentTipText()
          Returns the tip text for this property
 java.lang.String randomSeedTipText()
          Returns the tip text for this property
 java.lang.String replaceMissingValuesTipText()
          Returns the tip text for this property
 void setDistribution(SelectedTag newDstr)
          Sets the distribution to use for calculating the random matrix
 boolean setInputFormat(Instances instanceInfo)
          Sets the format of the input instances.
 void setNumberOfAttributes(int newAttNum)
          Sets the number of attributes (dimensions) the data should be reduced to
 void setOptions(java.lang.String[] options)
          Parses a given list of options.
 void setPercent(double newPercent)
          Sets the percent the attributes (dimensions) of the data should be reduced to
 void setRandomSeed(long seed)
          Sets the random seed of the random number generator
 void setReplaceMissingValues(boolean t)
          Sets either to use replace missing values filter or not
 
Methods inherited from class weka.filters.Filter
batchFilterFile, filterFile, getCapabilities, getOutputFormat, isFirstBatchDone, isNewBatch, isOutputFormatDefined, makeCopies, makeCopy, numPendingOutput, output, outputPeek, toString, useFilter, wekaStaticWrapper
 
Methods inherited from class java.lang.Object
equals, getClass, hashCode, notify, notifyAll, wait, wait, wait
 

Field Detail

SPARSE1

public static final int SPARSE1
distribution type: sparse 1

See Also:
Constant Field Values

SPARSE2

public static final int SPARSE2
distribution type: sparse 2

See Also:
Constant Field Values

GAUSSIAN

public static final int GAUSSIAN
distribution type: gaussian

See Also:
Constant Field Values

TAGS_DSTRS_TYPE

public static final Tag[] TAGS_DSTRS_TYPE
The types of distributions that can be used for calculating the random matrix

Constructor Detail

RandomProjection

public RandomProjection()
Method Detail

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 <number>
  The number of dimensions (attributes) the data should be reduced to
  (default 10; exclusive of the class attribute, if it is set).
 -D [SPARSE1|SPARSE2|GAUSSIAN]
  The distribution to use for calculating the random matrix.
  Sparse1 is:
    sqrt(3)*{-1 with prob(1/6), 0 with prob(2/3), +1 with prob(1/6)}
  Sparse2 is:
    {-1 with prob(1/2), +1 with prob(1/2)}
 
 -P <percent>
  The percentage of dimensions (attributes) the data should
  be reduced to (exclusive of the class attribute, if it is set). This -N
  option is ignored if this option is present or is greater
  than zero.
 -M
  Replace missing values using the ReplaceMissingValues filter
 -R <num>
  The random seed for the random number generator used for
  calculating the random matrix (default 42).

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 filter.

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

globalInfo

public java.lang.String globalInfo()
Returns a string describing this filter

Returns:
a description of the filter 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

numberOfAttributesTipText

public java.lang.String numberOfAttributesTipText()
Returns the tip text for this property

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

setNumberOfAttributes

public void setNumberOfAttributes(int newAttNum)
Sets the number of attributes (dimensions) the data should be reduced to

Parameters:
newAttNum - the goal for the dimensions

getNumberOfAttributes

public int getNumberOfAttributes()
Gets the current number of attributes (dimensionality) to which the data will be reduced to.

Returns:
the number of dimensions

percentTipText

public java.lang.String percentTipText()
Returns the tip text for this property

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

setPercent

public void setPercent(double newPercent)
Sets the percent the attributes (dimensions) of the data should be reduced to

Parameters:
newPercent - the percentage of attributes

getPercent

public double getPercent()
Gets the percent the attributes (dimensions) of the data will be reduced to

Returns:
the percentage of attributes

randomSeedTipText

public java.lang.String randomSeedTipText()
Returns the tip text for this property

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

setRandomSeed

public void setRandomSeed(long seed)
Sets the random seed of the random number generator

Parameters:
seed - the random seed value

getRandomSeed

public long getRandomSeed()
Gets the random seed of the random number generator

Returns:
the random seed value

distributionTipText

public java.lang.String distributionTipText()
Returns the tip text for this property

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

setDistribution

public void setDistribution(SelectedTag newDstr)
Sets the distribution to use for calculating the random matrix

Parameters:
newDstr - the distribution to use

getDistribution

public SelectedTag getDistribution()
Returns the current distribution that'll be used for calculating the random matrix

Returns:
the current distribution

replaceMissingValuesTipText

public java.lang.String replaceMissingValuesTipText()
Returns the tip text for this property

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

setReplaceMissingValues

public void setReplaceMissingValues(boolean t)
Sets either to use replace missing values filter or not

Parameters:
t - if true then the replace missing values is used

getReplaceMissingValues

public boolean getReplaceMissingValues()
Gets the current setting for using ReplaceMissingValues filter

Returns:
true if the replace missing values filter is used

getCapabilities

public Capabilities getCapabilities()
Returns the Capabilities of this filter.

Specified by:
getCapabilities in interface CapabilitiesHandler
Overrides:
getCapabilities in class Filter
Returns:
the capabilities of this object
See Also:
Capabilities

setInputFormat

public boolean setInputFormat(Instances instanceInfo)
                       throws java.lang.Exception
Sets the format of the input instances.

Overrides:
setInputFormat in class Filter
Parameters:
instanceInfo - an Instances object containing the input instance structure (any instances contained in the object are ignored - only the structure is required).
Returns:
true if the outputFormat may be collected immediately
Throws:
java.lang.Exception - if the input format can't be set successfully

input

public boolean input(Instance instance)
              throws java.lang.Exception
Input an instance for filtering.

Overrides:
input in class Filter
Parameters:
instance - the input instance
Returns:
true if the filtered instance may now be collected with output().
Throws:
java.lang.IllegalStateException - if no input format has been set
java.lang.NullPointerException - if the input format has not been defined.
java.lang.Exception - if the input instance was not of the correct format or if there was a problem with the filtering.

batchFinished

public boolean batchFinished()
                      throws java.lang.Exception
Signify that this batch of input to the filter is finished.

Overrides:
batchFinished in class Filter
Returns:
true if there are instances pending output
Throws:
java.lang.NullPointerException - if no input structure has been defined,
java.lang.Exception - if there was a problem finishing the batch.

getRevision

public java.lang.String getRevision()
Returns the revision string.

Specified by:
getRevision in interface RevisionHandler
Returns:
the revision

main

public static void main(java.lang.String[] argv)
Main method for testing this class.

Parameters:
argv - should contain arguments to the filter: use -h for help