prajna.text

Class BrillTagger

java.lang.Object

prajna.text.BrillTagger

All Implemented Interfaces:

PartOfSpeechTagger

public class BrillTagger
extends java.lang.Object
implements PartOfSpeechTagger

Implementation of a Brill Tagger, based on the algorithms of Eric Brill. For more information on the algorithm, see http://en.wikipedia.org/wiki/Brill_tagger

Author:

Edward Swing

Constructor Summary

Constructors

Constructor and Description
BrillTagger() Create a new instance of a Brill Tagger.
BrillTagger(java.lang.String model) Create a new instance of a Brill Tagger using the specified model path.

Method Summary

Methods

Modifier and Type	Method and Description
java.util.List<PartOfSpeechTag>	parseSentence(java.lang.String sentence) Parse the sentence and determine its part-of-speech tags
void	setModel(java.lang.String model) Set the model for this tagger.

Methods inherited from class java.lang.Object

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

Constructor Detail

BrillTagger

public BrillTagger()

Create a new instance of a Brill Tagger. The model - the path to the lexicon and rules files - must be set using setModel()

BrillTagger

public BrillTagger(java.lang.String model)

Create a new instance of a Brill Tagger using the specified model path. See setModel() for a description of the model path format

Parameters:

model - the model, or directory path with file names

Method Detail

parseSentence

public java.util.List<PartOfSpeechTag> parseSentence(java.lang.String sentence)

Parse the sentence and determine its part-of-speech tags

Specified by:

parseSentence in interface PartOfSpeechTagger

Parameters:

sentence - the sentence to parse

Returns:

a list of PartOfSpeechTag objects representing the sentence.

setModel

public void setModel(java.lang.String model)

Set the model for this tagger. The format for the model string should be: DirectoryPath;LexiconFileName;ContextualRuleFile;LexicalRuleFile. The directory indicated by the directory path should include all three files. If the model indicates only the directory, then the tagger looks for files named LEXICON, CONTEXTUALRULESFILE, and LEXICALRULESFILE in the directory.

Specified by:

setModel in interface PartOfSpeechTagger

Parameters:

model - the model, or directory path with file names