dragon.ir.search.smooth

Interface Smoother

All Known Implementing Classes:

AbsoluteDiscountSmoother, AbstractSmoother, DirichletSmoother, DocFirstTransSmoother, JMSmoother, OkapiSmoother, PivotedNormSmoother, QueryFirstTransSmoother, TFIDFSmoother, TwoStageSmoother

public interface Smoother

Interface of smoother which returns a score of a searching term in the given document to the searcher

The toolkit has implemented various language model smoothing methods as well as traditional probabilistic and vector space models. For language models, the score means the probability of the doucment generating the term.

Copyright: Copyright (c) 2005

Company: IST, Drexel University

Version:

1.0

Author:

Davis Zhou

Method Summary

Modifier and Type	Method and Description
boolean	getLogLikelihoodOption()
double	getSmoothedProb(int termFreq) Before calling this method, one should call the setQueryTerm method and the setDoc method.
double	getSmoothedProb(IRDoc doc) It is equal to calling getSmoothedProb(doc, 0);
double	getSmoothedProb(IRDoc doc, int termFreq) Before calling this method, one should call the setQueryTerm method.
double	getSmoothedProb(IRDoc doc, SimpleTermPredicate queryTerm) This method is equal to call getSmoothedProb(doc, queryTerm, 0).
double	getSmoothedProb(IRDoc doc, SimpleTermPredicate queryTerm, int termFreq)
double	getSmoothedProb(SimpleTermPredicate queryTerm) It is equal to calling getSmoothedProb(queryTerm, 0);
double	getSmoothedProb(SimpleTermPredicate queryTerm, int termFreq) Before calling this method, one should call the setDoc method.
boolean	isDocFirstOptimal() If this method returns true, the fullrank searcher will do breadth-frist search, i.e.
boolean	isQueryTermFirstOptimal() If this method returns true, the fullrank searcher will do depth-frist search, i.e.
void	setDoc(IRDoc doc) Set the current document for processing
void	setLogLikelihoodOption(boolean option) If this option is true, smoomther will return a log score
boolean	setParameters(double[] params)
void	setQueryTerm(SimpleTermPredicate queryTerm) Set the current query term for processing

Method Detail

getSmoothedProb

double getSmoothedProb(IRDoc doc,
                       SimpleTermPredicate queryTerm,
                       int termFreq)

Parameters:

doc - the document

queryTerm - the query term

termFreq - the frequency of the query term in the given document

Returns:

a score

getSmoothedProb

double getSmoothedProb(IRDoc doc,
                       SimpleTermPredicate queryTerm)

This method is equal to call getSmoothedProb(doc, queryTerm, 0).

Parameters:

doc - the document

queryTerm - the query term

Returns:

a score

getSmoothedProb

double getSmoothedProb(IRDoc doc,
                       int termFreq)

Before calling this method, one should call the setQueryTerm method.

Parameters:

doc - the document

termFreq - the frequency of the current query term in the given document

Returns:

a score

getSmoothedProb

double getSmoothedProb(SimpleTermPredicate queryTerm,
                       int termFreq)

Before calling this method, one should call the setDoc method.

Parameters:

queryTerm - the query term

termFreq - the frequency of the given query term in the current document

Returns:

a score

getSmoothedProb

double getSmoothedProb(int termFreq)

Before calling this method, one should call the setQueryTerm method and the setDoc method.

Parameters:

termFreq - the frequency of the current term in the current document

Returns:

a score

getSmoothedProb

double getSmoothedProb(SimpleTermPredicate queryTerm)

It is equal to calling getSmoothedProb(queryTerm, 0);

Parameters:

queryTerm - the query term

Returns:

a score

getSmoothedProb

double getSmoothedProb(IRDoc doc)

It is equal to calling getSmoothedProb(doc, 0);

Parameters:

doc - the document

Returns:

a score

setParameters

boolean setParameters(double[] params)

Parameters:

params - paramteres for the current smoother

Returns:

true if successful

setQueryTerm

void setQueryTerm(SimpleTermPredicate queryTerm)

Set the current query term for processing

Parameters:

queryTerm - the query term

setDoc

void setDoc(IRDoc doc)

Set the current document for processing

Parameters:

doc - the document

isDocFirstOptimal

boolean isDocFirstOptimal()

If this method returns true, the fullrank searcher will do breadth-frist search, i.e. processing document one by one.

Returns:

true or false

isQueryTermFirstOptimal

boolean isQueryTermFirstOptimal()

If this method returns true, the fullrank searcher will do depth-frist search, i.e. processing query terms one by one and then merging the document lists resulted from different query terms.

Returns:

true or false

setLogLikelihoodOption

void setLogLikelihoodOption(boolean option)

If this option is true, smoomther will return a log score

Parameters:

option - true or false

getLogLikelihoodOption

boolean getLogLikelihoodOption()