com.rapidminer.example
Class AttributeWeights

java.lang.Object
  com.rapidminer.operator.AbstractIOObject
      com.rapidminer.operator.ResultObjectAdapter
          com.rapidminer.tools.math.AverageVector
              com.rapidminer.example.AttributeWeights

All Implemented Interfaces:

IOObject, ResultObject, Saveable, LoggingHandler, java.io.Serializable, java.lang.Cloneable, java.lang.Comparable

public class AttributeWeights
extends AverageVector

AttributeWeights holds the information about the weights of attributes of an example set. It is delivered by several feature weighting algorithms or learning schemes. The use of a linked hash map ensures that the added features are stored in the same sequence they were added.

Author:

Ingo Mierswa Exp $

See Also:

Serialized Form

Field Summary

static int	ABSOLUTE_WEIGHTS Indicates that the the absolute weights should be used for sorting.
static int	DECREASING Indicates that the weights should be sorted in descending order.
static int	INCREASING Indicates that the weights should be sorted in ascending order.
static int	NO_SORTING Indicates that the weights should not be sorted at all.
static int	ORIGINAL_WEIGHTS Indicates that the the actual weights should be used for sorting.

Constructor Summary

AttributeWeights()
Creates a new empty attribute weights object.

AttributeWeights(ExampleSet exampleSet)
Creates a new attribute weights object containing a weight of 1 for each of the given input attributes.

Method Summary

java.lang.Object	clone() Returns a deep clone of the attribute weights which provides the same sequence of attribute names.
int	compareTo(java.lang.Object o) Since this average vector cannot be compared this method always returns 0.
DataTable	createDataTable()
boolean	equals(java.lang.Object o) Returns true if both objects have the same weight map.
java.util.Set<java.lang.String>	getAttributeNames() Returns an set of attribute names in this map ordered by their insertion time.
java.lang.String	getExtension() Returns the default extension in file choosers.
java.lang.String	getFileDescription() Returns the file description used in file choosers.
java.lang.String	getName() Returns the name of this AverageVector.
javax.swing.Icon	getResultIcon() Returns null.
int	getSortingType() Returns the currently used sorting type.
java.awt.Component	getVisualizationComponent(IOContainer container) Returns a visualisation component which allows sorting of the attribute weights and several weight plots.
double	getWeight(java.lang.String name) Returns the weight for the attribute with the given name.
int	getWeightType() Returns the currently used weight type.
int	hashCode() Returns the hash code of the weight map.
static AttributeWeights	load(java.io.File file) Loads a new AttributeWeights object from the given XML file.
void	normalize() This method normalizes all weights to the range 0 to 1.
void	save(java.io.File file) Saves the attribute weights into an XML file.
void	setSortingType(int sortingType) Sets the currently used sorting type.
void	setWeight(java.lang.String name, double weight) Sets the weight for the attribute with the given name.
void	setWeightType(int weightType) Returns the currently used weight type.
int	size() Returns the number of features in this map.
void	sortByWeight(java.lang.String[] attributeNames, int direction, int comparatorType) Sorts the given array of attribute names according to their weight, the sorting direction (ascending or descending), and with respect to the fact if original or absolute weights should be used.
java.lang.String	toString() Returns a string representation of this object.
void	writeAttributeWeights(java.io.File file, java.nio.charset.Charset encoding)

Methods inherited from class com.rapidminer.tools.math.AverageVector

addAveragable, buildAverages, getAveragable, getAveragable, getSize, removeAveragable, toResultString

Methods inherited from class com.rapidminer.operator.ResultObjectAdapter

addAction, getActions, isSavable, log, logError, logNote, logWarning, toHTML

Methods inherited from class com.rapidminer.operator.AbstractIOObject

copy, getLog, getSource, initWriting, read, setLoggingHandler, setSource, write

Methods inherited from class java.lang.Object

finalize, getClass, notify, notifyAll, wait, wait, wait

Methods inherited from interface com.rapidminer.operator.IOObject

copy, getLog, getSource, setLoggingHandler, setSource, write

Field Detail

NO_SORTING

public static final int NO_SORTING

Indicates that the weights should not be sorted at all.

See Also:

Constant Field Values

DECREASING

public static final int DECREASING

Indicates that the weights should be sorted in descending order.

See Also:

Constant Field Values

INCREASING

public static final int INCREASING

Indicates that the weights should be sorted in ascending order.

See Also:

Constant Field Values

ORIGINAL_WEIGHTS

public static final int ORIGINAL_WEIGHTS

Indicates that the the actual weights should be used for sorting.

See Also:

Constant Field Values

ABSOLUTE_WEIGHTS

public static final int ABSOLUTE_WEIGHTS

Indicates that the the absolute weights should be used for sorting.

See Also:

Constant Field Values

Constructor Detail

AttributeWeights

public AttributeWeights()

Creates a new empty attribute weights object.

AttributeWeights

public AttributeWeights(ExampleSet exampleSet)

Creates a new attribute weights object containing a weight of 1 for each of the given input attributes.

Method Detail

getName

public java.lang.String getName()

Returns the name of this AverageVector.

Specified by:

getName in interface ResultObject

Overrides:

getName in class ResultObjectAdapter

setWeight

public void setWeight(java.lang.String name,
                      double weight)

Sets the weight for the attribute with the given name.

getWeight

public double getWeight(java.lang.String name)

Returns the weight for the attribute with the given name. Returns Double.NaN if the weight for the queried attribute is not known.

getWeightType

public int getWeightType()

Returns the currently used weight type.

setWeightType

public void setWeightType(int weightType)

Returns the currently used weight type.

getSortingType

public int getSortingType()

Returns the currently used sorting type.

setSortingType

public void setSortingType(int sortingType)

Sets the currently used sorting type.

size

public int size()

Returns the number of features in this map.

Overrides:

size in class AverageVector

getAttributeNames

public java.util.Set<java.lang.String> getAttributeNames()

Returns an set of attribute names in this map ordered by their insertion time.

compareTo

public int compareTo(java.lang.Object o)

Since this average vector cannot be compared this method always returns 0.

equals

public boolean equals(java.lang.Object o)

Returns true if both objects have the same weight map.

Overrides:

equals in class AverageVector

hashCode

public int hashCode()

Returns the hash code of the weight map.

Overrides:

hashCode in class AverageVector

sortByWeight

public void sortByWeight(java.lang.String[] attributeNames,
                         int direction,
                         int comparatorType)

Sorts the given array of attribute names according to their weight, the sorting direction (ascending or descending), and with respect to the fact if original or absolute weights should be used.

Parameters:

direction - ASCENDING or DESCENDING

comparatorType - WEIGHT or WEIGHT_ABSOLUTE.

save

public void save(java.io.File file)
          throws java.io.IOException

Saves the attribute weights into an XML file.

Specified by:

save in interface Saveable

Overrides:

save in class ResultObjectAdapter

Throws:

java.io.IOException

writeAttributeWeights

public void writeAttributeWeights(java.io.File file,
                                  java.nio.charset.Charset encoding)
                           throws java.io.IOException

Throws:

java.io.IOException

load

public static AttributeWeights load(java.io.File file)
                             throws java.io.IOException

Loads a new AttributeWeights object from the given XML file.

Throws:

java.io.IOException

getExtension

public java.lang.String getExtension()

Description copied from interface: Saveable

Returns the default extension in file choosers.

getFileDescription

public java.lang.String getFileDescription()

Description copied from interface: Saveable

Returns the file description used in file choosers.

toString

public java.lang.String toString()

Returns a string representation of this object.

Overrides:

toString in class AverageVector

clone

public java.lang.Object clone()

Returns a deep clone of the attribute weights which provides the same sequence of attribute names.

Specified by:

clone in class AverageVector

normalize

public void normalize()

This method normalizes all weights to the range 0 to 1.

getVisualizationComponent

public java.awt.Component getVisualizationComponent(IOContainer container)

Returns a visualisation component which allows sorting of the attribute weights and several weight plots.

Specified by:

getVisualizationComponent in interface ResultObject

Overrides:

getVisualizationComponent in class AverageVector

getResultIcon

public javax.swing.Icon getResultIcon()

Description copied from class: ResultObjectAdapter

Returns null. Subclasses might want to override this method and returns an appropriate icon.

Specified by:

getResultIcon in interface ResultObject

Overrides:

getResultIcon in class ResultObjectAdapter

createDataTable

public DataTable createDataTable()