IBM Information Integrator for Content V8.2 APIs

com.ibm.mm.sdk.common.infomining.analysis
Class DKIKFClusterer

java.lang.Object
  |
  +--com.ibm.mm.sdk.common.infomining.analysis.DKIKFClusterer

public class DKIKFClusterer
extends java.lang.Object

A cluster can be used to partition a document collection into sets of similar documents. You can use only one clusterer object with the same service object. If you create a second clusterer object for the same service, all previously analyzed documents get lost and both clusterer objects represent the same clusterer.


Constructor Summary
DKIKFClusterer(DKIKFService ikfService)
          Creates a new clusterer object.
 
Method Summary
 void analyze(DKIKFTextDocument doc)
          Adds a document to the set of documents to be clustered.
 DKIKFClusterResult cluster()
          Clusters all documents sent to analyze(DKIKFTextDocument).
 int getClusterFeatureCount()
          Returns the number of cluster features to be returned.
 int getMaxClusterCount()
          Returns the maximum number of clusters to be returned.
 int getMinClusterCount()
          Returns the minimum number of clusters to be returned.
 DKIKFService getService()
          Returns the current service.
 void setClusterFeatureCount(int clusterFeatureCount)
          Sets the number of cluster features to be returned.
 void setMaxClusterCount(int maxClusterCount)
          Sets the maximum number of clusters to be returned.
 void setMinClusterCount(int minClusterCount)
          Sets the minimum number of clusters to be returned.
 
Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
 

Constructor Detail

DKIKFClusterer

public DKIKFClusterer(DKIKFService ikfService)
Creates a new clusterer object.
Parameters:
ikfService - a service object
Throws:
DKIKFAuthorizationException - if the user or group does not have the privilege IKFRunAnalysisFunc
Method Detail

getService

public DKIKFService getService()
Returns the current service.
Returns:
the current service

setMinClusterCount

public void setMinClusterCount(int minClusterCount)
Sets the minimum number of clusters to be returned.
Parameters:
minClusterCount - the minimum number of clusters to be returned
See Also:
getMinClusterCount()

getMinClusterCount

public int getMinClusterCount()
Returns the minimum number of clusters to be returned.
Returns:
the minimum number of clusters to be returned
See Also:
setMinClusterCount(int)

setMaxClusterCount

public void setMaxClusterCount(int maxClusterCount)
Sets the maximum number of clusters to be returned.
Parameters:
maxClusterCount - the maximum number of clusters to be returned
See Also:
getMaxClusterCount()

getMaxClusterCount

public int getMaxClusterCount()
Returns the maximum number of clusters to be returned.
Returns:
the maximum number of clusters to be returned
See Also:
setMaxClusterCount(int)

setClusterFeatureCount

public void setClusterFeatureCount(int clusterFeatureCount)
Sets the number of cluster features to be returned.
Returns:
the number of cluster features to be returned
See Also:
getClusterFeatureCount()

getClusterFeatureCount

public int getClusterFeatureCount()
Returns the number of cluster features to be returned.
Returns:
the number of cluster features to be returned
See Also:
setClusterFeatureCount(int)

analyze

public void analyze(DKIKFTextDocument doc)
             throws DKIKFUnsupportedLanguageException,
                    DKIKFDocumentAlreadyAnalyzedException,
                    DKIKFDocumentProcessingException
Adds a document to the set of documents to be clustered.
Parameters:
doc - a document to be added for clustering
Throws:
DKIKFDocumentLanguageNotSetException - if the specified document has no language
DKIKFDocumentNameNotSetException - if the specified document has no name
DKIKFUnsupportedLanguageException - if the document language is not supported by the clusterer
DKIKFDocumentAlreadyAnalyzedException - if a document with the same name has already been analyzed
DKIKFDocumentProcessingException - if an error occours during document processing
See Also:
cluster()

cluster

public DKIKFClusterResult cluster()
                           throws DKIKFClusteringException
Clusters all documents sent to analyze(DKIKFTextDocument).
Returns:
the cluster result
Throws:
DKIKFClusteringException - if an error occours during processing

IBM Information Integrator for Content V8.2 APIs

© Copyright International Business Machines Corporation 1996, 2003 IBM Corp. All rights reserved.