Uses of Interface
de.fu_berlin.ties.classify.Classifier

Packages that use Classifier
de.fu_berlin.ties.classify This package provides functionality for classification of texts and feature vectors. 
de.fu_berlin.ties.classify.winnow This package contains the Winnow classification algorithm and related algorithms and classes. 
de.fu_berlin.ties.extract This package handles information extraction and entitiy recognition. 
 

Uses of Classifier in de.fu_berlin.ties.classify
 

Classes in de.fu_berlin.ties.classify that implement Classifier
 class ExternalClassifier
          A proxy that provides a trainable classifier by communicating with an external (non-Java) program.
 class TrainableClassifier
          Classifiers extending this abstract class must provide a training mechanism by implementing the TrainableClassifier.doTrain(FeatureVector, String) method.
 

Uses of Classifier in de.fu_berlin.ties.classify.winnow
 

Classes in de.fu_berlin.ties.classify.winnow that implement Classifier
 class UltraconservativeWinnow
          A combination of Winnow with the "ultraconservative" approach proposed by Koby Crammer and Yoram Singer.
 class Winnow
          Classifier implementing the Winnow algorithm (Nick Littlestone).
 

Uses of Classifier in de.fu_berlin.ties.extract
 

Methods in de.fu_berlin.ties.extract that return Classifier
 Classifier ExtractorBase.getClassifier()
          Returns the classifier used for the local classification decisions.
 

Constructors in de.fu_berlin.ties.extract with parameters of type Classifier
ExtractorBase(String outExt, TargetStructure targetStruct, Classifier theClassifier, Representation theRepresentation, CombinationStrategy combiStrat, TokenizerFactory tFactory, TiesConfiguration config)
          Creates a new instance.
Extractor(String outExt, TargetStructure targetStruct, Classifier theClassifier, Representation theRepresentation, CombinationStrategy combiStrat, TokenizerFactory tFactory, TiesConfiguration config)
          Creates a new instance.
 



Copyright © 2003-2004 Christian Siefkes. All Rights Reserved.