Main Page | Class Hierarchy | Class List | Class Members

kddml.Core.DataMining.Clustering.DistributionBasedCluster Class Reference

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

kddml.Core.DataMining.Clustering.Cluster kddml.Core.DataMining.Clustering.DistributionBasedClusterManager kddml.Core.DataMining.Clustering.ClusterManager kddml.Core.DataMining.Clustering.ClusterManager List of all members.

Public Member Functions

 DistributionBasedCluster (String name, ClusterStatistic distribution)
boolean isDistributionBased ()
boolean isCentroidBased ()
Object getSeedAsInstance () throws ClusteringModelException
double[] getSeedCoordinate () throws ClusteringModelException
String toString ()

Protected Member Functions

Element toXML ()

Detailed Description

Model-based or distribution-based clustering methods assume the data of each cluster conforms to a specific statistical distribution (e.g. the Gaussian distribution) and the whole dataset is a mixture of several distribution models. EM is an example of distribution-based partitioning clustering that do not require the specification of distance measures.
A distribution-based cluster is always defined by their statistical information.

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.DistributionBasedCluster.DistributionBasedCluster String  name,
ClusterStatistic  distribution
 

Constructor given the cluster name and the statistic distribution.

Parameters:
name String
distribution ClusterStatistic


Member Function Documentation

boolean kddml.Core.DataMining.Clustering.DistributionBasedCluster.isDistributionBased  ) 
 

Returns true if the clustering is distribution-based.

Returns:
boolean true

Implements kddml.Core.DataMining.Clustering.ClusterManager.

boolean kddml.Core.DataMining.Clustering.DistributionBasedCluster.isCentroidBased  ) 
 

Returns true if the clustering is center-based.

Returns:
boolean false

Implements kddml.Core.DataMining.Clustering.ClusterManager.

Object kddml.Core.DataMining.Clustering.DistributionBasedCluster.getSeedAsInstance  )  throws ClusteringModelException [virtual]
 

Returns the representative cluster seed as instance. Returns the mean value of the distribution for a continuous field. Returns the most probable category of the distribution for a discrete field.

Exceptions:
ClusteringModelException 
Returns:
Object a single instance as in weka.core.Instance.

Implements kddml.Core.DataMining.Clustering.ClusterManager.

double [] kddml.Core.DataMining.Clustering.DistributionBasedCluster.getSeedCoordinate  )  throws ClusteringModelException [virtual]
 

Returns the seed coordinate as array of normalized doubles in respect on the ClusterDescription fields. Returns the normalized mean value of the distribution for a continuous field. Returns the normalized distribution for a discrete field.

Exceptions:
ClusteringModelException 
Returns:
double[]

Implements kddml.Core.DataMining.Clustering.ClusterManager.

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

Returns a representation of this cluster as string.

Returns:
String

Reimplemented from kddml.Core.DataMining.Clustering.Cluster.

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

Returns a representation of this cluster as PMML element.

Returns:
Element

Reimplemented from kddml.Core.DataMining.Clustering.Cluster.


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