Main Page | Class Hierarchy | Class List | Class Members

kddml.Operators.Preprocessing.PP_NORMALIZATION_SETTINGS Class Reference

Inheritance diagram for kddml.Operators.Preprocessing.PP_NORMALIZATION_SETTINGS:

kddml.Operators.Preprocessing.PPAlgorithmLoaderSettings kddml.Operators.AlgorithmLoaderSettings kddml.Operators.OperatorSettings List of all members.

Public Member Functions

String[] listAttributes ()
Boolean isRequired (String attribute_name)
Boolean isFixed (String attribute_name)
String getDefaultAttributeValue (String attribute_name)
KDDMLScalarType getAttributeType (String attribute_name)

Detailed Description

Normalization is the process of scaling data values of a numeric attribute to fit in a range such as [-1, 1] or [0,1]. Normalization is particularly useful for classification algorithms and clustering. The operator takes a preprocessing table, a list of numeric attributes, and a normalization method and returns a new preprocessing table with normalized values for each specified attribute. There are many methods for data discretization, each of them expressed by using the XML element ALGORITHM.
Attributes:


Member Function Documentation

String [] kddml.Operators.Preprocessing.PP_NORMALIZATION_SETTINGS.listAttributes  )  [virtual]
 

Returns the list of attributes belonging to the operator specification.

Returns:
String[]

Reimplemented from kddml.Operators.AlgorithmLoaderSettings.

Boolean kddml.Operators.Preprocessing.PP_NORMALIZATION_SETTINGS.isRequired String  attribute_name  )  [virtual]
 

Returns true if a given attribute is required in the operator specification. Returns false if the attribute is implied or fixed. Finally, the method returns null if the attribute name do not belong to the list of attributes returned by the method listAttributes().

Parameters:
attribute_name String
Returns:
boolean

Reimplemented from kddml.Operators.AlgorithmLoaderSettings.

Boolean kddml.Operators.Preprocessing.PP_NORMALIZATION_SETTINGS.isFixed String  attribute_name  )  [virtual]
 

Returns true if a given attribute is fixed (i.e. constant) in the operator specification. Returns false if the attribute is implied or required. Finally, the method returns null if the attribute name do not belong to the list of attributes returned by the method listAttributes().

Parameters:
attribute_name String
Returns:
boolean

Reimplemented from kddml.Operators.AlgorithmLoaderSettings.

String kddml.Operators.Preprocessing.PP_NORMALIZATION_SETTINGS.getDefaultAttributeValue String  attribute_name  )  [virtual]
 

Returns the default value related to a given input attribute. Returns null if the attribute do not admit a default value. An attribute can be given any legal value as a default. The attribute value is not required on each element in the document, but if it is not present, it will appear to be the specified default. If the attribute is implied and it is not included in the element, the operator assumes that this is the attribute value.

Parameters:
attribute_name String
Returns:
String

Reimplemented from kddml.Operators.AlgorithmLoaderSettings.

KDDMLScalarType kddml.Operators.Preprocessing.PP_NORMALIZATION_SETTINGS.getAttributeType String  attribute_name  )  [virtual]
 

Returns the type related to a given input attribute. KDDML supports the attribute types as expressed by the class KDDMLScalarType. Returns null if the attribute name do no belong to the list of attributes returned by the method listAttributes().

Parameters:
attribute_name String
Returns:
KDDMLScalarType

Reimplemented from kddml.Operators.AlgorithmLoaderSettings.


Generated on Thu Feb 23 13:04:53 2006 for kddml by  doxygen 1.4.3