fr.lip6.jkernelmachines.active

Class SimpleAL<T>

java.lang.Object

fr.lip6.jkernelmachines.active.ActiveLearner<T>

fr.lip6.jkernelmachines.active.SimpleAL<T>

public class SimpleAL<T>
extends ActiveLearner<T>

Simple active learning strategy as presented in:
Support vector machine active learning with applications to text classification. S. Tong and D. Koller. Journal of Machine Learning Research, 2:45–66, 2001.

Author:

picard

Field Summary

Fields inherited from class fr.lip6.jkernelmachines.active.ActiveLearner

classifier, train

Constructor Summary

Constructors

Constructor and Description
SimpleAL(OnlineClassifier<T> c, java.util.List<TrainingSample<T>> l)

Method Summary

Modifier and Type	Method and Description
TrainingSample<T>	getActiveSample(java.util.List<TrainingSample<T>> l) Method returning the best training sample in the list with respect to the active strategy
void	updateClassifier(int nbSamples) perform nbSample updates of the classifier using the active strategy

Methods inherited from class fr.lip6.jkernelmachines.active.ActiveLearner

getClassifier, getTrain, setClassifier, setTrain

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

SimpleAL

public SimpleAL(OnlineClassifier<T> c,
                java.util.List<TrainingSample<T>> l)

Method Detail

getActiveSample

public TrainingSample<T> getActiveSample(java.util.List<TrainingSample<T>> l)

Description copied from class: ActiveLearner

Method returning the best training sample in the list with respect to the active strategy

Specified by:

getActiveSample in class ActiveLearner<T>

Returns:

the best sample

updateClassifier

public void updateClassifier(int nbSamples)

Description copied from class: ActiveLearner

perform nbSample updates of the classifier using the active strategy

Specified by:

updateClassifier in class ActiveLearner<T>