dragon.nlp.tool.xtract

Class EngWordPairExpand

java.lang.Object

dragon.nlp.tool.xtract.EngWordPairExpand

All Implemented Interfaces:

WordPairExpand

public class EngWordPairExpand
extends Object
implements WordPairExpand

Expanding word pair (not necessary consecutive) to noun phrase

Copyright: Copyright (c) 2005

Company: IST, Drexel University

Version:

1.0

Author:

Davis Zhou

Field Summary

Fields

Modifier and Type	Field and Description
protected IntSparseMatrix[]	arrPairSentLeftMatrix
protected IntSparseMatrix[]	arrPairSentRightMatrix
protected String	indexFolder
protected int	maxSpan
protected IntSparseMatrix	sentMatrix
protected double	threshold
protected SimpleElementList	wordList

Constructor Summary

Constructors

Constructor and Description
EngWordPairExpand(int maxSpan, String indexFolder, double threshold)

Method Summary

Modifier and Type	Method and Description
protected boolean	checkEndingWordValidation(String word, int posIndex)
protected Token	checkSentPos(int spanFromFirstWord, ArrayList sentList)
protected boolean	checkStartingWordValidation(String word, int posIndex)
protected boolean	checkValidation(String word, int posIndex)
ArrayList	expand(WordPairStat wordPairStat, int span) Expand the input word pair to a real phrase
protected Token	expandSecion(int start, int end, int sentNum, boolean inverse, int direction, ArrayList sentList)
protected ArrayList	getSentenceList(WordPairStat wordPairStat, int span)
protected String	getWordContent(int index)

Methods inherited from class java.lang.Object

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

Field Detail

sentMatrix

protected IntSparseMatrix sentMatrix

wordList

protected SimpleElementList wordList

maxSpan

protected int maxSpan

indexFolder

protected String indexFolder

arrPairSentLeftMatrix

protected IntSparseMatrix[] arrPairSentLeftMatrix

arrPairSentRightMatrix

protected IntSparseMatrix[] arrPairSentRightMatrix

threshold

protected double threshold

Constructor Detail

EngWordPairExpand

public EngWordPairExpand(int maxSpan,
                         String indexFolder,
                         double threshold)

Method Detail

expand

public ArrayList expand(WordPairStat wordPairStat,
                        int span)

Description copied from interface: WordPairExpand

Expand the input word pair to a real phrase

Specified by:

expand in interface WordPairExpand

Parameters:

wordPairStat - the word pair object

span - the distance between two words

Returns:

a list of Token objects. Each token contain the extracted phrase and its frequency in the corpus

getSentenceList

protected ArrayList getSentenceList(WordPairStat wordPairStat,
                                    int span)

expandSecion

protected Token expandSecion(int start,
                             int end,
                             int sentNum,
                             boolean inverse,
                             int direction,
                             ArrayList sentList)

checkSentPos

protected Token checkSentPos(int spanFromFirstWord,
                             ArrayList sentList)

getWordContent

protected String getWordContent(int index)

checkValidation

protected boolean checkValidation(String word,
                                  int posIndex)

checkEndingWordValidation

protected boolean checkEndingWordValidation(String word,
                                            int posIndex)

checkStartingWordValidation

protected boolean checkStartingWordValidation(String word,
                                              int posIndex)