Main Page | Class Hierarchy | Class List | Class Members

kddml.Operators.Preprocessing.PP_HIERARCHICAL_DISCRETIZATION_RESOLVER Class Reference

Inheritance diagram for kddml.Operators.Preprocessing.PP_HIERARCHICAL_DISCRETIZATION_RESOLVER:

kddml.Operators.Preprocessing.InstanceLevelIndependentTransformation kddml.Operators.Preprocessing.PreprocessingResolver kddml.Operators.OperatorResolver kddml.Operators.HWResourcesDescription List of all members.

Public Member Functions

void readAttributes (Hashtable< String, KDDMLScalarManager > parameters) throws ResolverException, KDDMLCoreException
KDDMLObjectType getArgumentType (int i)
int estimatedRAMUsage ()

Protected Member Functions

void readOthersArgurments (Vector< KDDMLObject > arguments) throws KDDMLCoreException, ResolverException
void readDataStatistics (DataStatisticsManager stat, DataStatisticsManager PPstat) throws ResolverException, KDDMLCoreException
String getHistoryDescription ()
Instance[] runCore (Instance tupla, Instance metatupla) throws ResolverException, KDDMLCoreException

Detailed Description

A categoric discretization of an attribute with respect to a hierarchy is performed. Given a nonnumeric attribute of the input preprocessing table, a concept hierarchy related to the attribute values, and an integer representing the hierarchy level, the operator returns a new table containing generalized values for that attribute. The mapping between attribute values and leaf values of the hierarchy is by name.
If the attribute value is found in a leaf of the hierarchy, then the generalized node value is computed at the specified level. In this case, the old attribute value is replaced with the generalized item that has been found.
If some value cannot be found in the hierarchy leaves or if the generalization cannot be computed, then the attribute is set with a missing value for that instance. Also the correspondent preprocessing information is marked with the exception no hierarchy generalization. Obviously, the level of the hierarchy must be compatible with the depth of the hierarchy in order to compute the generalization.

Title: KDDML

Description: Knowledge Discovery in Database Environment

Copyright: Copyright (c) 2003-2005

Company: Universita' di Pisa - Dipartimento di Informatica

Author:
Andrea Romei (romei@di.unipi.it)

Sandra Zimei

Version:
2.0.16


Member Function Documentation

void kddml.Operators.Preprocessing.PP_HIERARCHICAL_DISCRETIZATION_RESOLVER.readAttributes Hashtable< String, KDDMLScalarManager parameters  )  throws ResolverException, KDDMLCoreException [virtual]
 

Reads the XML attributes related to a generic preprocessing operator. An operator settings object captures the attributes associated with a particular operator. It allows a knowledgeable user to fine tune operator parameters. Generally, not all parameters must be specified, however, those specified are taken into account by the KDDML.
Attributes are given as hashtable, where the key is the name of the attribute related to the operator and the value is a KDDMLScalar object containing the attribute value. Attribute value is checked by the interpreter layer and its type is correct.

Parameters:
parameters Hashtable the attributes related to the operator. The key of the hashtable is the name of the operator. The value of the hashtable is a KDDMLScalar representing the value of the operator.
Exceptions:
ResolverException if a resolving error occurs.
KDDMLCoreException if a level core error occurs.

Implements kddml.Operators.Preprocessing.PreprocessingResolver.

void kddml.Operators.Preprocessing.PP_HIERARCHICAL_DISCRETIZATION_RESOLVER.readOthersArgurments Vector< KDDMLObject arguments  )  throws KDDMLCoreException, ResolverException [protected]
 

Reads other arguments from the input vector, i.e. arguments whose position is different from the first. By default, this method do nothing. It must be overried if and only if the preprocessing operator accepts other arguments in addiction to the preprocessing table. In this case, also the method getArgumentType(int i) must be overried.

Parameters:
arguments Vector the list of arguments as KDDMLObject.
Exceptions:
KDDMLCoreException if a level core error occurs.
ResolverException if a level operator error occurs.

Reimplemented from kddml.Operators.Preprocessing.PreprocessingResolver.

void kddml.Operators.Preprocessing.PP_HIERARCHICAL_DISCRETIZATION_RESOLVER.readDataStatistics DataStatisticsManager  stat,
DataStatisticsManager  PPstat
throws ResolverException, KDDMLCoreException [protected]
 

Reads data statistics related to input preprocessing table. Data statistic can be used to provide additional information to preprocessing operator, such as the number of total instances or attributes belonging to the data source. By default, this method do nothing. It can be overried inside operator implementation if necessary.

Parameters:
stat DataStatisticsManager statistics related to physical instances
PPstat DataStatisticsManager statistics related to preprocessing instances
Exceptions:
ResolverException if a resolving error occurs.
KDDMLCoreException if a level core error occurs.

Reimplemented from kddml.Operators.Preprocessing.PreprocessingResolver.

String kddml.Operators.Preprocessing.PP_HIERARCHICAL_DISCRETIZATION_RESOLVER.getHistoryDescription  )  [protected, virtual]
 

Returns a description of the actions performed by this preprocessing operator. This description will be reported in the history related to the preprocessing data source.

Returns:
String
Exceptions:
KDDMLCoreException 

Implements kddml.Operators.Preprocessing.PreprocessingResolver.

Instance [] kddml.Operators.Preprocessing.PP_HIERARCHICAL_DISCRETIZATION_RESOLVER.runCore Instance  tupla,
Instance  metatupla
throws ResolverException, KDDMLCoreException [protected, virtual]
 

Core operator method. Given a physical instance and the related preprocessing tuple as weka.core.Instance, the operator returns the modified instances as two-dimensional array.

Parameters:
tupla Instance the physical instance on wich the operator works.
metatupla Instance the preprocessing instance on wich the operator works.
Returns:
Instance[] a two-dimensional array containing the calculated output instance. The first element contains the physical instaces; the second element contains the related preprocessing instances.
Exceptions:
ResolverException if a resolving error occurs.
KDDMLCoreException if a level core error occurs.

Implements kddml.Operators.Preprocessing.InstanceLevelIndependentTransformation.

KDDMLObjectType kddml.Operators.Preprocessing.PP_HIERARCHICAL_DISCRETIZATION_RESOLVER.getArgumentType int  i  )  [virtual]
 

Returns the expected type of the i-th argument of the operator. Returns null if the operator do not admit argument as i-th child.

Parameters:
i int
Returns:
KDDMLObjectType

Reimplemented from kddml.Operators.Preprocessing.PreprocessingResolver.

int kddml.Operators.Preprocessing.PP_HIERARCHICAL_DISCRETIZATION_RESOLVER.estimatedRAMUsage  ) 
 

The estimated RAM usage for the operator. Must be a value between 0 and 1024 MB.

Returns:
boolean

Reimplemented from kddml.Operators.OperatorResolver.


Generated on Thu Feb 23 13:04:52 2006 for kddml by  doxygen 1.4.3