Main Page | Class Hierarchy | Class List | Class Members

kddml.Core.DataMining.Clustering.Cluster Class Reference

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

kddml.Core.DataMining.Clustering.ClusterManager kddml.Core.DataMining.Clustering.CentroidBasedCluster kddml.Core.DataMining.Clustering.DistributionBasedCluster List of all members.

Public Member Functions

 Cluster (String name)
int getClusterId ()
String getName ()
Integer getSize ()
void setSize (int num_elements)
void setCovariancesMatrix (CategoryMatrixManager matrix)
CategoryMatrixManager getCovariancesMatrix ()
ClusterStatisticManager getClusterStatistic ()
String toString ()
void setClusterStatistic (ClusterStatisticManager statistics)

Protected Member Functions

double normalizeFieldValue (String field_name, double value) throws ClusteringModelException
double[] normalizeFieldValue (String field_name, String category) throws ClusteringModelException
Element toXML ()

Protected Attributes

ClusterDescription centroid_description
int identifier
String name
Integer size
ClusterStatistic cluster_distribution
CategoryMatrix covariances_matrix

Detailed Description

A Cluster object holds the metadata about a cluster discovered by running a clustering algorithm. Clusters are always associated with a ClusteringModel. A cluster is defined by its center vector or by statistics. This information is contained in a ClusterDescription. The definition of a cluster may contain a center vector as well as statistics and the cluster is defined by the centroid (for center-based) od by statistics (for distribution-based). Each cluster is identified by an identifier, that is automatically generated by the constructor. A covariace matrix can be given for this cluster that contain information on overall data distribution,. A covariance matrix stores coordinate-by-coordinate variances (diagonal cells) and covariances (non-diagonal cells).

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.Cluster.Cluster String  name  ) 
 

Constructor given the cluster name.

Parameters:
name String


Member Function Documentation

int kddml.Core.DataMining.Clustering.Cluster.getClusterId  ) 
 

Returns the cluster id.

Returns:
int

Implements kddml.Core.DataMining.Clustering.ClusterManager.

String kddml.Core.DataMining.Clustering.Cluster.getName  ) 
 

Returns the name of the cluster designated by the clustering algorithm.

Returns:
String

Implements kddml.Core.DataMining.Clustering.ClusterManager.

Integer kddml.Core.DataMining.Clustering.Cluster.getSize  ) 
 

Returns the number of cases in the portion of the training data assigned to the cluster during the model build. This is a non-negative value. It can be null.

Returns:
Integer

Implements kddml.Core.DataMining.Clustering.ClusterManager.

void kddml.Core.DataMining.Clustering.Cluster.setSize int  num_elements  ) 
 

Sets the number of cases in the portion of the training data assigned to the cluster during the model build. This must be a non-negative value.

Parameters:
num_elements int

Implements kddml.Core.DataMining.Clustering.ClusterManager.

void kddml.Core.DataMining.Clustering.Cluster.setCovariancesMatrix CategoryMatrixManager  matrix  ) 
 

Sets the covariance matrix.

Parameters:
matrix CategoryMatrix

Implements kddml.Core.DataMining.Clustering.ClusterManager.

CategoryMatrixManager kddml.Core.DataMining.Clustering.Cluster.getCovariancesMatrix  ) 
 

Returns the covariances matrix provided for this cluster. Returns null if no covariances matrix is avaible.

Returns:
CategoryMatrixManager the convariance matrix. Can be null.

Implements kddml.Core.DataMining.Clustering.ClusterManager.

ClusterStatisticManager kddml.Core.DataMining.Clustering.Cluster.getClusterStatistic  ) 
 

Returns the cluster distribution. Can be null, for example for center-based clustering.

Returns:
ClusterStatisticManager

Implements kddml.Core.DataMining.Clustering.ClusterManager.

double kddml.Core.DataMining.Clustering.Cluster.normalizeFieldValue String  field_name,
double  value
throws ClusteringModelException [protected]
 

Normalizes the input value for a numeric field.

Parameters:
field_name String
value double
Exceptions:
ClusteringModelException 
Returns:
double

double [] kddml.Core.DataMining.Clustering.Cluster.normalizeFieldValue String  field_name,
String  category
throws ClusteringModelException [protected]
 

Normalizes the input value for a discrete field.

Parameters:
field_name String
category String
Exceptions:
ClusteringModelException 
Returns:
double[]

String kddml.Core.DataMining.Clustering.Cluster.toString  ) 
 

Returns a representation of this cluster as string.

Returns:
String

Reimplemented in kddml.Core.DataMining.Clustering.CentroidBasedCluster, and kddml.Core.DataMining.Clustering.DistributionBasedCluster.

Element kddml.Core.DataMining.Clustering.Cluster.toXML  )  [protected]
 

Returns a representation of this cluster as PMML element.

Returns:
Element

Reimplemented in kddml.Core.DataMining.Clustering.CentroidBasedCluster, and kddml.Core.DataMining.Clustering.DistributionBasedCluster.

void kddml.Core.DataMining.Clustering.Cluster.setClusterStatistic ClusterStatisticManager  statistics  ) 
 

Sets the cluster statistics. This field is optional for center-based clustering and it is required for distribution-based clustering.

Parameters:
statistics ClusterStatisticManager

Implements kddml.Core.DataMining.Clustering.ClusterManager.


Member Data Documentation

ClusterDescription kddml.Core.DataMining.Clustering.Cluster.centroid_description [protected]
 

The centroid description for this cluster. Setted in ClusteringModel. Equal for each cluster.

int kddml.Core.DataMining.Clustering.Cluster.identifier [protected]
 

The unique identifier of the cluster.

String kddml.Core.DataMining.Clustering.Cluster.name [protected]
 

The cluster name.

Integer kddml.Core.DataMining.Clustering.Cluster.size [protected]
 

The number of elements. Can be null.

ClusterStatistic kddml.Core.DataMining.Clustering.Cluster.cluster_distribution [protected]
 

The cluster statistic. This field is optional.

CategoryMatrix kddml.Core.DataMining.Clustering.Cluster.covariances_matrix [protected]
 

The covariances matrix used to store variances (diagonal cells) and covariances (non-diagonal cells).


Generated on Thu Feb 23 13:04:40 2006 for kddml by  doxygen 1.4.3