dragon.ir.summarize

Class LexRankSummarizer

java.lang.Object

dragon.ir.summarize.AbstractSentenceSum

dragon.ir.summarize.LexRankSummarizer

All Implemented Interfaces:

GenericMultiDocSummarizer

Direct Known Subclasses:

ClusterLexRankSummarizer, SemanticRankSummarizer

public class LexRankSummarizer
extends AbstractSentenceSum
implements GenericMultiDocSummarizer

LexRank Summerizer

Copyright: Copyright (c) 2005

Company: IST, Drexel University

Version:

1.0

Author:

Davis Zhou

Field Summary

Fields

Modifier and Type	Field and Description
protected CollectionReader	collectionReader
protected DocDistance	distanceMetric
protected OnlineSentenceIndexer	indexer
protected OnlineSentenceIndexReader	indexReader
protected PowerMethod	powerMethod
protected double	threshold
protected boolean	useContinuousValue
protected boolean	useTFIDF

Constructor Summary

Constructors

Constructor and Description
LexRankSummarizer(OnlineSentenceIndexer indexer)
LexRankSummarizer(OnlineSentenceIndexer indexer, boolean useTFIDF)

Method Summary

Modifier and Type	Method and Description
protected DoubleDenseMatrix	buildWeightMatrix(ArrayList docSet)
protected double	computeSimilarity(IRDoc firstSent, IRDoc secondSent) LexRank Summarizer uses cosine similrity.
protected ArrayList	getSentenceSet(IndexReader indexReader)
void	setContinuousScoreOpiton(boolean option)
void	setSimilarityThreshold(double threshold)
String	summarize(CollectionReader collectionReader, int maxLength)

Methods inherited from class dragon.ir.summarize.AbstractSentenceSum

buildSummary, buildSummary

Methods inherited from class java.lang.Object

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

Field Detail

indexReader

protected OnlineSentenceIndexReader indexReader

indexer

protected OnlineSentenceIndexer indexer

collectionReader

protected CollectionReader collectionReader

distanceMetric

protected DocDistance distanceMetric

powerMethod

protected PowerMethod powerMethod

threshold

protected double threshold

useContinuousValue

protected boolean useContinuousValue

useTFIDF

protected boolean useTFIDF

Constructor Detail

LexRankSummarizer

public LexRankSummarizer(OnlineSentenceIndexer indexer)

LexRankSummarizer

public LexRankSummarizer(OnlineSentenceIndexer indexer,
                         boolean useTFIDF)

Method Detail

setSimilarityThreshold

public void setSimilarityThreshold(double threshold)

setContinuousScoreOpiton

public void setContinuousScoreOpiton(boolean option)

summarize

public String summarize(CollectionReader collectionReader,
                        int maxLength)

Specified by:

summarize in interface GenericMultiDocSummarizer

Parameters:

collectionReader - the collection reader

maxLength - the length of the summary

Returns:

the summary in natural language.

buildWeightMatrix

protected DoubleDenseMatrix buildWeightMatrix(ArrayList docSet)

computeSimilarity

protected double computeSimilarity(IRDoc firstSent,
                                   IRDoc secondSent)

LexRank Summarizer uses cosine similrity. This method can be overrided.

Parameters:

firstSentIndex: - the number of the first sentence

secondSentIndex: - the number of the second sentence

Returns:

the similarity between two specified sentences. The similarity value ranges from 0 to 1.

getSentenceSet

protected ArrayList getSentenceSet(IndexReader indexReader)