edu.ucla.sspace.wordsi.psd

Class PseudoWordContextExtractor

java.lang.Object

edu.ucla.sspace.wordsi.psd.PseudoWordContextExtractor

All Implemented Interfaces:

ContextExtractor

public class PseudoWordContextExtractor
extends Object
implements ContextExtractor

A pseudo word based ContextExtractor. A mapping from real tokens to pseudo words is used to automatically replace tokens in a corpus while Wordsi processes contexts. Only pseudo words are represented in the Wordsi space. When a token is encountered, if it has a pseudo word mapping, that instance is replaced with the pseudo word mapping. A context vector will be generated for the context surrounded that word instance, and the pseudo word replacement will serve as the primary key for the reporter and the raw token will serve as the secondary key. The pseudo word will then replace the raw token in the context for all other words, and thus serve as a feature in place of the real token.

Author:

Keith Stevens

Constructor Summary

Constructors

Constructor and Description
PseudoWordContextExtractor(ContextGenerator generator, int windowSize, Map<String,String> pseudoWordMap) Creates a new PseudoWordContextExtracto.

Method Summary

Methods

Modifier and Type	Method and Description
int	getVectorLength() Returns the maximum number of dimensions used to represent any given context.
void	processDocument(BufferedReader document, Wordsi wordsi) Processes the content of document and calls Wordsi.handleContextVector(java.lang.String, java.lang.String, edu.ucla.sspace.vector.SparseDoubleVector) for each context vector that can be extracted from document.

Methods inherited from class java.lang.Object

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

Constructor Detail

PseudoWordContextExtractor

public PseudoWordContextExtractor(ContextGenerator generator,
                          int windowSize,
                          Map<String,String> pseudoWordMap)

Creates a new PseudoWordContextExtracto.

Parameters:

generator - The ContextGenerator responsible for creating context vectors

windowSize - The number of words before and after the focus word which compose a context

pseudoWordMap - The mapping from real words to their pseudo word replacements

Method Detail

getVectorLength

public int getVectorLength()

Returns the maximum number of dimensions used to represent any given context.

Specified by:

getVectorLength in interface ContextExtractor

processDocument

public void processDocument(BufferedReader document,
                   Wordsi wordsi)

Processes the content of document and calls Wordsi.handleContextVector(java.lang.String, java.lang.String, edu.ucla.sspace.vector.SparseDoubleVector) for each context vector that can be extracted from document.

Specified by:

processDocument in interface ContextExtractor