Main Page | Class Hierarchy | Class List | Class Members

kddml.Operators.DataMining.ClusteringAlgorithms.EM_RESOLVER Class Reference

Inheritance diagram for kddml.Operators.DataMining.ClusteringAlgorithms.EM_RESOLVER:

kddml.Operators.DataMining.ClusteringAlgorithms.ClusteringAlgorithmResolverTask kddml.Operators.DataMining.DMAlgorithmResolverTask kddml.Operators.AlgorithmResolverTask List of all members.

Public Member Functions

void readParameters (Hashtable< String, KDDMLScalarManager > parameters) throws ResolverException, KDDMLCoreException
void readXMLDestination (String xml_dest)
Object inputFormatting (InternalTableManager instances) throws ResolverException, KDDMLCoreException
Object execute (Object instances) throws ResolverException, KDDMLCoreException
ClusteringModel outputFormatting (Object clusters) throws ResolverException, KDDMLCoreException
String[] getTmpFiles ()

Detailed Description

The EM (Expectation Maximization) algorithm is an effective, popular technique for estimating mixture model parameters (cluster parameters and their mixture weights). The EM algorithm iteratively refines initial mixture model parameter estimates to better fit the data and it terminates at a locally optimal solution. EM is a distribution-based technique with the euclidean distance as comparison measure. The field comparison function si the ABS function for numeric attributes and the DELTA function for discrete attributes.
The EM algorithm is implemented using the WEKA library and it supports nominal or continuous attributes only: no string attributes are allowed in the input table.

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


Member Function Documentation

void kddml.Operators.DataMining.ClusteringAlgorithms.EM_RESOLVER.readParameters Hashtable< String, KDDMLScalarManager parameters  )  throws ResolverException, KDDMLCoreException
 

Reads the XML parameters related to a generic algorithm stored in the ALGORITHM entity. An algorithm settings object captures the parameters associated with a particular algorithm. It allows a knowledgeable user to fine tune algorithm parameters. Generally, not all parameters must be specified, however, those specified are taken into account by the KDDML.
Parameters are given as hashtable, where the key is the name of the parameter related to the algorithm and the value is a KDDMLScalar object containing the parameter value. Parameter value is checked by the interpreter layer and its type is correct.

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

Implements kddml.Operators.AlgorithmResolverTask.

void kddml.Operators.DataMining.ClusteringAlgorithms.EM_RESOLVER.readXMLDestination String  xml_dest  ) 
 

Reads the XML destination that stores the final result.

Parameters:
xml_dest String

Implements kddml.Operators.DataMining.DMAlgorithmResolverTask.

Object kddml.Operators.DataMining.ClusteringAlgorithms.EM_RESOLVER.inputFormatting InternalTableManager  instances  )  throws ResolverException, KDDMLCoreException [virtual]
 

Formatting of the input dataset. It scans the passed data in order to transform the data source into the format as the algorithm specification requires. The method returns a generic java.lang.Object (e.g. the name of the temporary file created during the formatting process) that is used as input of the excute() method.

Parameters:
instances InternalTableManager the data set to scan.
Returns:
Object the set of weka.core.Instances instances representing the data source.
Exceptions:
ResolverException if a resolving error occurs.
KDDMLCoreException if a level core error occurs.

Implements kddml.Operators.DataMining.DMAlgorithmResolverTask.

Object kddml.Operators.DataMining.ClusteringAlgorithms.EM_RESOLVER.execute Object  instances  )  throws ResolverException, KDDMLCoreException [virtual]
 

Main execute method. It takes the result of the inputFormatting() method and, tipically, the procedure calls an external DM library implemented in C,C++ in order to extract the model. The method returns a generic java.lang.Object (e.g. the name of the temporary file created during the extraction process) that is used as input of the outputFormatting(Object obj) method.

Parameters:
instances the set of weka.core.Instances instances representing the data source.
Returns:
Object th weka.clusterers.EM representing the set of clusters.
Exceptions:
ResolverException if a resolving error occurs.
KDDMLCoreException if a level core error occurs.

Implements kddml.Operators.DataMining.DMAlgorithmResolverTask.

ClusteringModel kddml.Operators.DataMining.ClusteringAlgorithms.EM_RESOLVER.outputFormatting Object  clusters  )  throws ResolverException, KDDMLCoreException [virtual]
 

Interprets the output generated by the algorithm in order to return an appropriate ClusteringModel. The method takes as input a generic java.lang.Object that is the result of the execute() method.

Parameters:
clusters Object the weka.clusterers.EM representing the set of clusters.
Returns:
ClusteringModel the final result of the extraction process.
Exceptions:
ResolverException if a resolving error occurs.
KDDMLCoreException if a level core error occurs.

Implements kddml.Operators.DataMining.ClusteringAlgorithms.ClusteringAlgorithmResolverTask.

String [] kddml.Operators.DataMining.ClusteringAlgorithms.EM_RESOLVER.getTmpFiles  ) 
 

Returns the list of temporary files used during the mining. Files will be removed at the end of execution of the algorithm. Returns an empty array if the algorithm do not generate temporary files.

Returns:
String[]

Implements kddml.Operators.DataMining.DMAlgorithmResolverTask.


Generated on Thu Feb 23 13:04:48 2006 for kddml by  doxygen 1.4.3