edu.stanford.nlp.classify
Class NaiveBayesClassifier

java.lang.Object
  |
  +--edu.stanford.nlp.classify.AbstractClassifier
        |
        +--edu.stanford.nlp.classify.NaiveBayesClassifier

All Implemented Interfaces:

Classifier

public class NaiveBayesClassifier

extends AbstractClassifier

Naive Bayes Classifier.

Field Summary

Fields inherited from class edu.stanford.nlp.classify.AbstractClassifier

testmatrix, trainlabels, trainmatrix

Constructor Summary

NaiveBayesClassifier()

Method Summary

protected int	assignCategory(int datum) assigns datum datum to a category and returns the index of the category.
protected Array	assignMultipleCategories(int datum) assigns datum datum to multiple categories with certain probabilities.
Array	classify(LabeledDataCollection train, DataCollection test) Assigns each Datum in the test collection to one single category.
protected Array	create_pr_s(Matrix cm) creates pr_s given the count matrix C(w,s)
protected Matrix	create_pr_w_s(Matrix cm) Creates pr_w_s given the count matrix C(w,s)
protected double	score(int category, int datum) gives the Naive Bayes score of a given sense (category) for a given datum in test set
Matrix	softClassify(LabeledDataCollection train, DataCollection test) Assigns each Datum in the test collection to multiple categories with certain probabilities.

Methods inherited from class edu.stanford.nlp.classify.AbstractClassifier

initialize

Methods inherited from class java.lang.Object

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

Constructor Detail

NaiveBayesClassifier

public NaiveBayesClassifier()

Method Detail

softClassify

public Matrix softClassify(LabeledDataCollection train,
                           DataCollection test)

Description copied from interface: Classifier

Assigns each Datum in the test collection to multiple categories with certain probabilities. Returns a Matrix whose columns are the category assignment distributions for each Datum.

Specified by:

softClassify in interface Classifier

Overrides:

softClassify in class AbstractClassifier

classify

public Array classify(LabeledDataCollection train,
                      DataCollection test)

Description copied from interface: Classifier

Assigns each Datum in the test collection to one single category. Returns an Array that gives datum_index-catogory_index pairs.

Specified by:

classify in interface Classifier

Overrides:

classify in class AbstractClassifier

create_pr_w_s

protected Matrix create_pr_w_s(Matrix cm)

Creates pr_w_s given the count matrix C(w,s)

create_pr_s

protected Array create_pr_s(Matrix cm)

creates pr_s given the count matrix C(w,s)

score

protected double score(int category,
                       int datum)

gives the Naive Bayes score of a given sense (category) for a given datum in test set

assignCategory

protected int assignCategory(int datum)

assigns datum datum to a category and returns the index of the category.

Specified by:

assignCategory in class AbstractClassifier

assignMultipleCategories

protected Array assignMultipleCategories(int datum)

assigns datum datum to multiple categories with certain probabilities. returns array of category_index-probability pairs.

Specified by:

assignMultipleCategories in class AbstractClassifier