Package org.terrier.matching.dsms

Class MRFDependenceScoreModifier

java.lang.Object

org.terrier.matching.dsms.DependenceScoreModifier

org.terrier.matching.dsms.MRFDependenceScoreModifier

All Implemented Interfaces:

java.lang.Cloneable, DocumentScoreModifier

public class MRFDependenceScoreModifier
extends DependenceScoreModifier

Implements Markov Random Fields. See Metzler & Croft, SIGIR 2005. Note that this implementation does not utilise the frequency of a tuple in the collection - instead, this is assumed to be a constant, as per the implementation in the Ivory retrieval system. Properties:

• See properties for DependenceScoreModifier
• mrf.mu - Mu of MRF model, in the Dirichlet model.

References: Metzler, D. and Croft, W.B., "A Markov Random Field Model for Term Dependencies," Proceedings of the 28th annual international ACM SIGIR conference on Research and development in information retrieval (SIGIR 2005), 472-479, 2005

Since:

3.0

Author:

Craig Macdonald

Field Summary

Fields

Modifier and Type	Field	Description
protected double	MU

Fields inherited from class org.terrier.matching.dsms.DependenceScoreModifier

avgDocLen, dependency, logger, ngramLength, numTokens, phraseQTWfnid, phraseTerms, w_o, w_t, w_u

Constructor Summary

Constructors

Constructor	Description
MRFDependenceScoreModifier()

Method Summary

Modifier and Type	Method	Description
boolean	modifyScores(Index index, MatchingQueryTerms terms, ResultSet set)	Modifies the scores of documents, in which there exist, or there does not exist a given phrase.
protected double	scoreFDSD(int matchingNGrams, int _docLength)
void	setCollectionStatistics(CollectionStatistics cs, Index _index)	Sets the collection statistics used to score the documents (number of documents in the collection, etc)

Methods inherited from class org.terrier.matching.dsms.DependenceScoreModifier

calculateDependence, clone, countTrue, determineGlobalStatistics, doDependency, getName, NOR, openPostingLists, score, scoreFDSD

Methods inherited from class java.lang.Object

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

Field Detail

MU

protected double MU

Constructor Detail

MRFDependenceScoreModifier

public MRFDependenceScoreModifier()

Method Detail

modifyScores

public boolean modifyScores(Index index,
                            MatchingQueryTerms terms,
                            ResultSet set)

Modifies the scores of documents, in which there exist, or there does not exist a given phrase.

Specified by:

modifyScores in interface DocumentScoreModifier

Overrides:

modifyScores in class DependenceScoreModifier

Parameters:

index - Index the data structures to use.

terms - MatchingQueryTerms the terms to be matched for the query. This does not correspond to the phrase terms necessarily, but to all the terms of the query.

set - ResultSet the result set for the query.

Returns:

true if any scores have been altered

scoreFDSD

protected double scoreFDSD(int matchingNGrams,
                           int _docLength)

Specified by:

scoreFDSD in class DependenceScoreModifier

setCollectionStatistics

public void setCollectionStatistics(CollectionStatistics cs,
                                    Index _index)

Sets the collection statistics used to score the documents (number of documents in the collection, etc)

Overrides:

setCollectionStatistics in class DependenceScoreModifier