Main Page | Class Hierarchy | Class List | Class Members

kddml.Core.DataMining.ClassificationTrees.ScoreDistribution Class Reference

Inheritance diagram for kddml.Core.DataMining.ClassificationTrees.ScoreDistribution:

kddml.Core.DataMining.ClassificationTrees.ScoreDistributionManager List of all members.

Public Member Functions

 ScoreDistribution (String[] categories, double[] number_of_instances) throws TreeModelException
 ScoreDistribution (String[] categories)
double getCount (String name)
double getPercCount (String name)
String getHighestCategory ()
double getHighestValue ()
double getTotal ()
int getNumberOfValues ()
Enumeration getCategories ()
String toString ()

Protected Member Functions

 ScoreDistribution (Vector elements)
boolean updateDistribution (String value, double record_count)

Detailed Description

The score distribution is a method to list predicted values in a classification trees structure. If the node holds an enumeration, each entry of the enumeration is stored in an hashtable structure where:
  1. the key is the category of the target attribute;
  2. the value is the size (i.e. the number of records) associated with the category
When a SimpleNode is selected as the final node and if this node has no 'score' attribute, then the highest recordCount the ScoreDistribution determines which value is selected as predicted class.

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)
Version:
2.0.16


Constructor & Destructor Documentation

kddml.Core.DataMining.ClassificationTrees.ScoreDistribution.ScoreDistribution String[]  categories,
double[]  number_of_instances
throws TreeModelException
 

Constructor.

Parameters:
categories String[] the list of categories for the target attribute.
number_of_instances double[] the number of records associated with the categories.
Exceptions:
TreeModelException if the size of categories and the size of number_of_instances differs.

kddml.Core.DataMining.ClassificationTrees.ScoreDistribution.ScoreDistribution String[]  categories  ) 
 

Constructor given the categories values for the target attribute. The number of records related to the category is 0.

Parameters:
categories String[] the list of categories for the target attribute.

kddml.Core.DataMining.ClassificationTrees.ScoreDistribution.ScoreDistribution Vector  elements  )  [protected]
 

Proprietary constructor given a list of PMML elements.

Parameters:
elements Vector


Member Function Documentation

double kddml.Core.DataMining.ClassificationTrees.ScoreDistribution.getCount String  name  ) 
 

Returns the record count of a given category. Returns -1 if no category can be found in the target attribute.

Parameters:
name String the category name
Returns:
double

Implements kddml.Core.DataMining.ClassificationTrees.ScoreDistributionManager.

double kddml.Core.DataMining.ClassificationTrees.ScoreDistribution.getPercCount String  name  ) 
 

Returns the record count as percentage of a given category. Returns -1 if no category can be found in the target attribute.

Parameters:
name String the category name
Returns:
double a value between 0 and 1 or -1 if the category is not found.

Implements kddml.Core.DataMining.ClassificationTrees.ScoreDistributionManager.

String kddml.Core.DataMining.ClassificationTrees.ScoreDistribution.getHighestCategory  ) 
 

Returns the highest category value of this score distribution. If two or more values have the highest distribution, then the method returns the first category value.

Returns:
String

Implements kddml.Core.DataMining.ClassificationTrees.ScoreDistributionManager.

double kddml.Core.DataMining.ClassificationTrees.ScoreDistribution.getHighestValue  ) 
 

Returns the highest record count of this score distribution.

Returns:
double

Implements kddml.Core.DataMining.ClassificationTrees.ScoreDistributionManager.

double kddml.Core.DataMining.ClassificationTrees.ScoreDistribution.getTotal  ) 
 

Returns the total sum of record counts.

Returns:
double a positive double

Implements kddml.Core.DataMining.ClassificationTrees.ScoreDistributionManager.

int kddml.Core.DataMining.ClassificationTrees.ScoreDistribution.getNumberOfValues  ) 
 

Returns the number of categories for this score distribution.

Returns:
int a positive integer

Implements kddml.Core.DataMining.ClassificationTrees.ScoreDistributionManager.

boolean kddml.Core.DataMining.ClassificationTrees.ScoreDistribution.updateDistribution String  value,
double  record_count
[protected]
 

Updates a distribution for a given category. Returns false if the value do not exist in the hashtable. Returns true if updating has success.

Parameters:
value String
record_count double
Returns:
boolean

Enumeration kddml.Core.DataMining.ClassificationTrees.ScoreDistribution.getCategories  ) 
 

Returns the list categories as enumeration.

Returns:
Enumeration list of strings.

Implements kddml.Core.DataMining.ClassificationTrees.ScoreDistributionManager.

String kddml.Core.DataMining.ClassificationTrees.ScoreDistribution.toString  ) 
 

Returns a representation of this object as string.

Returns:
String


Generated on Thu Feb 23 13:04:39 2006 for kddml by  doxygen 1.4.3