edu.harvard.seas.iis.abilities.classify
Class PositiveAndUnlabeledClassifier

java.lang.Object
  weka.classifiers.Classifier
      edu.harvard.seas.iis.abilities.classify.PositiveAndUnlabeledClassifier

All Implemented Interfaces:

java.io.Serializable, java.lang.Cloneable, weka.core.CapabilitiesHandler, weka.core.OptionHandler, weka.core.RevisionHandler

public class PositiveAndUnlabeledClassifier
extends weka.classifiers.Classifier

Author:

Charles Herrmann, kgajos This classifier learns from positive and unlabeled examples. It is an implementation of the methods described in Elkan, C., & Noto, K. (2008). Learning classifiers from only positive and unlabeled data. KDD '08: Proceeding of the 14th ACM SIGKDD international conference on Knowledge discovery and data mining.

See Also:

Serialized Form

Constructor Summary

PositiveAndUnlabeledClassifier()

PositiveAndUnlabeledClassifier(weka.classifiers.Classifier baseClassifier)

PositiveAndUnlabeledClassifier(weka.classifiers.Classifier baseClassifier, java.lang.String[] features)

Method Summary

void	buildClassifier(weka.core.Instances trainSet)
void	buildClassifier1(weka.core.Instances trainSet)
void	buildClassifier2(weka.core.Instances trainSet)
void	buildClassifier3(weka.core.Instances trainSet)
double	classifyInstance(weka.core.Instance point)
static PositiveAndUnlabeledClassifier	deserializeFromFile(java.io.File f)
double[]	distributionForInstance(weka.core.Instance inst)
java.lang.String[]	getAllowedFeatures()
double	getDeliberateProbability(weka.core.Instance point) Returns the probability that a particular movement (represented by the Instance) was deliberate
int	getMethodForEvaluatingC()
void	setAllowedFeatures(java.lang.String[] allowedFeatures)
void	setMethodForEvaluatingC(int methodForEvaluatingC)
java.lang.String	toString()

Methods inherited from class weka.classifiers.Classifier

debugTipText, forName, getCapabilities, getDebug, getOptions, getRevision, listOptions, makeCopies, makeCopy, setDebug, setOptions

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, wait, wait, wait

Constructor Detail

PositiveAndUnlabeledClassifier

public PositiveAndUnlabeledClassifier()
                               throws java.lang.Exception

Throws:

java.lang.Exception

PositiveAndUnlabeledClassifier

public PositiveAndUnlabeledClassifier(weka.classifiers.Classifier baseClassifier)

PositiveAndUnlabeledClassifier

public PositiveAndUnlabeledClassifier(weka.classifiers.Classifier baseClassifier,
                                      java.lang.String[] features)

Parameters:

baseClassifier - the underlying probabilistic classifier (See the paper for more details);

features - names of features that should be used by the classifier (this allows for some features to be ignored); this list has to include the "Class" feature

Method Detail

getMethodForEvaluatingC

public int getMethodForEvaluatingC()

setMethodForEvaluatingC

public void setMethodForEvaluatingC(int methodForEvaluatingC)

getAllowedFeatures

public java.lang.String[] getAllowedFeatures()

setAllowedFeatures

public void setAllowedFeatures(java.lang.String[] allowedFeatures)

classifyInstance

public double classifyInstance(weka.core.Instance point)
                        throws java.lang.Exception

Overrides:

classifyInstance in class weka.classifiers.Classifier

Throws:

java.lang.Exception

getDeliberateProbability

public double getDeliberateProbability(weka.core.Instance point)
                                throws java.lang.Exception

Returns the probability that a particular movement (represented by the Instance) was deliberate

Parameters:

point -

Returns:

Throws:

java.lang.Exception

distributionForInstance

public double[] distributionForInstance(weka.core.Instance inst)
                                 throws java.lang.Exception

Overrides:

distributionForInstance in class weka.classifiers.Classifier

Throws:

java.lang.Exception

buildClassifier

public void buildClassifier(weka.core.Instances trainSet)
                     throws java.lang.Exception

Specified by:

buildClassifier in class weka.classifiers.Classifier

Throws:

java.lang.Exception

buildClassifier1

public void buildClassifier1(weka.core.Instances trainSet)
                      throws java.lang.Exception

Throws:

java.lang.Exception

buildClassifier2

public void buildClassifier2(weka.core.Instances trainSet)
                      throws java.lang.Exception

Throws:

java.lang.Exception

buildClassifier3

public void buildClassifier3(weka.core.Instances trainSet)
                      throws java.lang.Exception

Throws:

java.lang.Exception

deserializeFromFile

public static PositiveAndUnlabeledClassifier deserializeFromFile(java.io.File f)
                                                          throws java.io.IOException,
                                                                 java.lang.ClassNotFoundException

Throws:

java.io.IOException

java.lang.ClassNotFoundException

toString

public java.lang.String toString()

Overrides:

toString in class java.lang.Object