uk.ac.gla.terrier.matching.models
Class Idf

java.lang.Object
  uk.ac.gla.terrier.matching.models.Idf

All Implemented Interfaces:

java.io.Serializable, java.lang.Cloneable

public final class Idf
extends java.lang.Object
implements java.io.Serializable, java.lang.Cloneable

This class computes the idf values for specific terms in the collection.

Author:

Gianni Amati, Ben He, Vassilis Plachouras

See Also:

Serialized Form

Field Summary

static double	LOG_2_OF_E The natural logarithm of 2, used to change the base of logarithms.
static double	REC_LOG_2_OF_E The reciprocal of CONSTANT, computed for efficiency.

Constructor Summary

Idf()
A default constructor.

Idf(double docs)
A constructor specifying the number of documents in the collection.

Method Summary

java.lang.Object	clone() Make a perfect clone of this object
double	idf(double d) Returns the idf of d.
double	idf(int d) Returns the idf of the given number d.
double	idfDFR(double d) Returns the idf of d.
double	idfDFR(int d) Returns the idf of the given number d.
double	idfENQUIRY(double d) The INQUERY idf formula.
double	idfN(double d) Return the normalised idf of the given number.
double	idfN(int d) Return the normalised idf of the given number.
double	idfNENQUIRY(double d) The normalised INQUERY idf formula
static double	log(double d) Returns the base 2 log of the given double precision number.
static double	log(double d1, double d2) Returns the base 2 log of d1 over d2
void	setNumberOfDocuments(double N)

Methods inherited from class java.lang.Object

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

Field Detail

LOG_2_OF_E

public static final double LOG_2_OF_E

The natural logarithm of 2, used to change the base of logarithms.

REC_LOG_2_OF_E

public static final double REC_LOG_2_OF_E

The reciprocal of CONSTANT, computed for efficiency.

Constructor Detail

Idf

public Idf()

A default constructor. NOTE: You must set the number of documents if you intend to use the idf* functions in this class

Idf

public Idf(double docs)

A constructor specifying the number of documents in the collection.

Parameters:

docs - The number of documents.

Method Detail

clone

public java.lang.Object clone()

Make a perfect clone of this object

Overrides:

clone in class java.lang.Object

idf

public final double idf(double d)

Returns the idf of d.

Parameters:

d - The given term frequency

Returns:

the base 2 log of numberOfDocuments/d

idf

public final double idf(int d)

Returns the idf of the given number d.

Parameters:

d - the number for which the idf will be computed.

Returns:

the idf of the given number d.

idfDFR

public final double idfDFR(double d)

Returns the idf of d.

Parameters:

d - The given term frequency

Returns:

the base 2 log of numberOfDocuments/d

idfDFR

public final double idfDFR(int d)

Returns the idf of the given number d.

Parameters:

d - the number for which the idf will be computed.

Returns:

the idf of the given number d.

idfENQUIRY

public final double idfENQUIRY(double d)

The INQUERY idf formula. We need to check again this formula, as it seems that there is a bug in the expression numberOfDocuments - d / d.

Parameters:

d - the number for which the idf will be computed

Returns:

the INQUERY idf of the number d

idfN

public final double idfN(double d)

Return the normalised idf of the given number.

Parameters:

d - The number of which the idf is computed.

Returns:

the normalised idf of d

setNumberOfDocuments

public void setNumberOfDocuments(double N)

idfN

public final double idfN(int d)

Return the normalised idf of the given number.

Parameters:

d - The number of which the idf is computed.

Returns:

the normalised idf of d

idfNENQUIRY

public final double idfNENQUIRY(double d)

The normalised INQUERY idf formula

Parameters:

d - the number for which we will compute the normalised idf

Returns:

the normalised INQUERY idf of d

log

public static final double log(double d)

Returns the base 2 log of the given double precision number.

Parameters:

d - The number of which the log we will compute

Returns:

the base 2 log of the given number

log

public static final double log(double d1,
                               double d2)

Returns the base 2 log of d1 over d2

Parameters:

d1 - the nominator

d2 - the denominator

Returns:

the base 2 log of d1/d2