org.apache.mahout.classifier.bayes
Interface Algorithm

All Known Implementing Classes:

BayesAlgorithm, CBayesAlgorithm

public interface Algorithm

The algorithm interface for implementing variations of bayes Algorithm

Method Summary

ClassifierResult	classifyDocument(String[] document, Datastore datastore, String defaultCategory) Classify the document and return the Result
ClassifierResult[]	classifyDocument(String[] document, Datastore datastore, String defaultCategory, int numResults) Classify the document and return the top numResults
double	documentWeight(Datastore datastore, String label, String[] document) Calculate the document weight as the dot product of document vector and the corresponding weight vector of a particular class
double	featureWeight(Datastore datastore, String label, String feature) Get the weighted probability of the feature.
Collection<String>	getLabels(Datastore datastore) Returns the labels in the given Model
void	initialize(Datastore datastore) Initialize the data store and verifies the data in it.

Method Detail

initialize

void initialize(Datastore datastore)
                throws InvalidDatastoreException

Initialize the data store and verifies the data in it.

Parameters:

datastore -

Throws:

InvalidDatastoreException

classifyDocument

ClassifierResult classifyDocument(String[] document,
                                  Datastore datastore,
                                  String defaultCategory)
                                  throws InvalidDatastoreException

Classify the document and return the Result

Parameters:

document - The document to classify

datastore - The data store(InMemory)

defaultCategory - The default category to assign Ties are broken by comparing the category

Returns:

A Collection of ClassifierResults.

Throws:

InvalidDatastoreException

classifyDocument

ClassifierResult[] classifyDocument(String[] document,
                                    Datastore datastore,
                                    String defaultCategory,
                                    int numResults)
                                    throws InvalidDatastoreException

Classify the document and return the top numResults

Parameters:

document - The document to classify

datastore - The Datastore (InMemory)

defaultCategory - The default category to assign

numResults - The maximum number of results to return, ranked by score. Ties are broken by comparing the category

Returns:

A Collection of ClassifierResults.

Throws:

InvalidDatastoreException

featureWeight

double featureWeight(Datastore datastore,
                     String label,
                     String feature)
                     throws InvalidDatastoreException

Get the weighted probability of the feature.

Parameters:

datastore - The Datastore (InMemory)

label - The label of the feature

feature - The feature to calc. the prob. for

Returns:

The weighted probability

Throws:

InvalidDatastoreException

documentWeight

double documentWeight(Datastore datastore,
                      String label,
                      String[] document)

Calculate the document weight as the dot product of document vector and the corresponding weight vector of a particular class

Parameters:

datastore - The Datastore (InMemory)

label - The label to calculate the probability of

document - The document

Returns:

The probability

See Also:

featureWeight(Datastore, String, String)

getLabels

Collection<String> getLabels(Datastore datastore)
                             throws InvalidDatastoreException

Returns the labels in the given Model

Parameters:

datastore - The Datastore (InMemory)

Returns:

Collection of labels

Throws:

InvalidDatastoreException