org.terrier.matching.models.normalisation

Class NormalisationB

java.lang.Object

org.terrier.matching.models.normalisation.Normalisation

org.terrier.matching.models.normalisation.NormalisationB

All Implemented Interfaces:

Serializable, Cloneable

public class NormalisationB
extends Normalisation

This class implements BM25's normalisation method.

Author:

Ben He

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
protected String	methodName The name of the normalisation method .

Fields inherited from class org.terrier.matching.models.normalisation.Normalisation

averageDocumentLength, idf, Nt, numberOfDocuments, numberOfTokens, parameter, termFrequency

Constructor Summary

Constructors

Constructor and Description
NormalisationB() The default constructor.

Method Summary

Methods

Modifier and Type	Method and Description
String	getInfo() Get the name of the normalisation method.
double	normalise(double tf, double docLength, double termFrequency) This method gets the normalised term frequency.

Methods inherited from class org.terrier.matching.models.normalisation.Normalisation

clone, getParameter, setAverageDocumentLength, setDocumentFrequency, setNumberOfDocuments, setNumberOfTokens, setParameter

Methods inherited from class java.lang.Object

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

Field Detail

methodName

protected final String methodName

The name of the normalisation method .

See Also:

Constant Field Values

Constructor Detail

NormalisationB

public NormalisationB()

The default constructor. The hyper-parameter value is set to 0.75 by default.

Method Detail

getInfo

public String getInfo()

Get the name of the normalisation method.

Specified by:

getInfo in class Normalisation

Returns:

Return the name of the normalisation method.

normalise

public double normalise(double tf,
               double docLength,
               double termFrequency)

This method gets the normalised term frequency.

Specified by:

normalise in class Normalisation

Parameters:

tf - The frequency of the query term in the document.

docLength - The number of tokens in the document.

termFrequency - The frequency of the query term in the collection.

Returns:

The normalised term frequency.