Uses of Class
de.fu_berlin.ties.classify.PredictionDistribution

Packages that use PredictionDistribution
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. 
 

Uses of PredictionDistribution in de.fu_berlin.ties.classify
 

Methods in de.fu_berlin.ties.classify that return PredictionDistribution
protected  PredictionDistribution ExternalClassifier.doClassify(FeatureVector features, Set candidateClasses)
          Classifies an item that is represented by a feature vector by choosing the most probable class among a set of candidate classes.
 PredictionDistribution TrainableClassifier.classify(FeatureVector features, Set candidateClasses)
          Classifies an item that is represented by a feature vector by choosing the most probable class among a set of candidate classes.
protected abstract  PredictionDistribution TrainableClassifier.doClassify(FeatureVector features, Set candidateClasses)
          Classifies an item that is represented by a feature vector by choosing the most probable class among a set of candidate classes.
 PredictionDistribution TrainableClassifier.trainOnError(FeatureVector features, String targetClass, Set candidateClasses)
          Handles error-driven learning ("train only errors"): the specified feature vector is trained into the model only if the predicted class for the feature vector differs from the specified target class.
 PredictionDistribution Classifier.classify(FeatureVector features, Set candidateClasses)
          Classifies an item that is represented by a feature vector by choosing the most probable class among a set of candidate classes.
 

Methods in de.fu_berlin.ties.classify with parameters of type PredictionDistribution
protected  boolean TrainableClassifier.trainOnErrorHook(PredictionDistribution predDist, FeatureVector features, String targetClass, Set candidateClasses)
          Subclasses can implement this hook for more refined error-driven learning.
 

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

Subclasses of PredictionDistribution in de.fu_berlin.ties.classify.winnow
 class WinnowDistribution
          A distribution over the classes predicted by a Winnow classifier.
 

Methods in de.fu_berlin.ties.classify.winnow that return PredictionDistribution
protected  PredictionDistribution Winnow.doClassify(FeatureVector features, Set candidateClasses)
          Classifies an item that is represented by a feature vector by choosing the most probable class among a set of candidate classes.
 

Methods in de.fu_berlin.ties.classify.winnow with parameters of type PredictionDistribution
protected  boolean Winnow.trainOnErrorHook(PredictionDistribution predDist, FeatureVector features, String targetClass, Set candidateClasses)
          Hook implementing error-driven learning, promoting and demoting weights as required.
 



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