Main Page | Class Hierarchy | Class List | Class Members

kddml.Core.DataMining.Clustering.SIMILARITY_MATRIX Class Reference

Inheritance diagram for kddml.Core.DataMining.Clustering.SIMILARITY_MATRIX:

kddml.Core.DataMining.Clustering.AttributeComparisonMeasure List of all members.

Public Member Functions

 SIMILARITY_MATRIX (SymmetricMatrixManager similarity_matrix)
String getFunctionName ()

Protected Member Functions

double computeNumeric (double value1, Double value2) throws ClusteringModelException
double computeDiscrete (String value1, String value2) throws ClusteringModelException

Detailed Description

Given two field values x and y, the similarity matrix stores coordinate-by-coordinate the similarity values. In this case comparison is a symmetric matrix which contains the similarity values for each discrete value. Similarity matrix can be used on discrete attributes only.

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.Clustering.SIMILARITY_MATRIX.SIMILARITY_MATRIX SymmetricMatrixManager  similarity_matrix  ) 
 

Constructor given the similarity matrix.

Parameters:
similarity_matrix SymmetricMatrixManager


Member Function Documentation

double kddml.Core.DataMining.Clustering.SIMILARITY_MATRIX.computeNumeric double  value1,
Double  value2
throws ClusteringModelException [protected, virtual]
 

Computes the comparison measure for a numeric attribute. Throws an exception if the function cannot be calculated on a numeric attribute.

Parameters:
center_coordinate double the centroid coordinate for this attribute. Cannot be null.
value2 Double the attribute value to compare. Can be null for missing attributes.
Exceptions:
ClusteringModelException if the function cannot be calculated.
Returns:
double

Implements kddml.Core.DataMining.Clustering.AttributeComparisonMeasure.

double kddml.Core.DataMining.Clustering.SIMILARITY_MATRIX.computeDiscrete String  value1,
String  value2
throws ClusteringModelException [protected, virtual]
 

Computes the comparison measure for a discrete attribute. Throws an exception if the function cannot be calculated on a discrete attribute.

Parameters:
value1 String the centroid coordinate for this attribute. Cannot be null.
value2 String the attribute value to compare. Can be null for missing attributes.
Exceptions:
ClusteringModelException if the function cannot be calculated.
Returns:
double

Implements kddml.Core.DataMining.Clustering.AttributeComparisonMeasure.

String kddml.Core.DataMining.Clustering.SIMILARITY_MATRIX.getFunctionName  )  [virtual]
 

Returns the function name.

Returns:
String

Implements kddml.Core.DataMining.Clustering.AttributeComparisonMeasure.


Generated on Thu Feb 23 13:04:41 2006 for kddml by  doxygen 1.4.3