org.terrier.matching.models

Class StaticScoreModifierWeightingModel

java.lang.Object

org.terrier.matching.models.WeightingModel

org.terrier.matching.models.StaticScoreModifierWeightingModel

All Implemented Interfaces:

Serializable, Cloneable, Model

Direct Known Subclasses:

StaticFeature

public abstract class StaticScoreModifierWeightingModel
extends WeightingModel

Base abstract class for query independent features loaded from file. For types of file are supported:

• oos/ois: an ObjectOutputStream file of float[] or double[] array
• docid2score: a text file containing [docid] [score]
• listofscores: a text file containing one score per line
• tmap: a TIntDoubleHashMap saved in an ObjectOutputStream file

Since:

4.0

Author:

Craig Macdonald

See Also:

Serialized Form

Field Summary

Fields inherited from class org.terrier.matching.models.WeightingModel

averageDocumentLength, c, cs, documentFrequency, es, i, keyFrequency, numberOfDocuments, numberOfPointers, numberOfTokens, numberOfUniqueTerms, rq, termFrequency

Method Summary

Methods

Modifier and Type	Method and Description
protected static double[]	castToDoubleArr(float[] f)
protected static double[]	castToDoubleArr(short[] f)
protected static float[]	castToFloatArr(double[] f)
protected static float[]	castToFloatArr(short[] f)
String	getInfo() Returns the name of the model.
double	getScoreD(int docid)
float	getScoreF(int docid)
String	getSource()
protected void	loadDocid2score(int numDocs, String inputFile, int column)
protected void	loadfloatOOS(String inputFile)
protected void	loadOOS(String inputFile)
protected void	loadScorefile(int numDocs, String inputFile, int column)
static float	max(float[] a) Return the max of the specified array
static float	min(float[] a) Return the min of the specified array
protected static void	printStats(double[] ar)
double	score(double tf, double docLength) This method provides the contract for implementing weighting models.
double	score(double tf, double docLength, double n_t, double F_t, double _keyFrequency) This method provides the contract for implementing weighting models.
abstract double	score(Posting p) Returns score
static float[]	standardNormalisation(float[] data) Normalises the data in the specified array to be in range [0,1], with 0 as the minimum, and 1 as the maximum.

Methods inherited from class org.terrier.matching.models.WeightingModel

clone, getOverflowed, getParameter, prepare, setCollectionStatistics, setEntryStatistics, setKeyFrequency, setParameter, setRequest

Methods inherited from class java.lang.Object

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

Method Detail

getScoreD

public final double getScoreD(int docid)

getScoreF

public final float getScoreF(int docid)

getSource

public final String getSource()

score

public abstract double score(Posting p)

Description copied from class: WeightingModel

Returns score

Overrides:

score in class WeightingModel

Returns:

score

getInfo

public String getInfo()

Description copied from class: WeightingModel

Returns the name of the model.

Specified by:

getInfo in interface Model

Specified by:

getInfo in class WeightingModel

Returns:

java.lang.String

score

public double score(double tf,
           double docLength)

Description copied from class: WeightingModel

This method provides the contract for implementing weighting models.

Specified by:

score in class WeightingModel

Parameters:

tf - The term frequency in the document

docLength - the document's length

Returns:

the score assigned to a document with the given tf and docLength, and other preset parameters

score

public double score(double tf,
           double docLength,
           double n_t,
           double F_t,
           double _keyFrequency)

Description copied from class: WeightingModel

This method provides the contract for implementing weighting models. As of Terrier 3.6, the 5-parameter score method is being deprecated since is is not used. The two parameter score method should be used instead. Tagged for removal in a later version.

Specified by:

score in class WeightingModel

Parameters:

tf - The term frequency in the document

docLength - the document's length

n_t - The document frequency of the term

F_t - the term frequency in the collection

_keyFrequency - the term frequency in the query

Returns:

the score returned by the implemented weighting model.

loadDocid2score

protected void loadDocid2score(int numDocs,
                   String inputFile,
                   int column)

loadScorefile

protected void loadScorefile(int numDocs,
                 String inputFile,
                 int column)

loadOOS

protected void loadOOS(String inputFile)

loadfloatOOS

protected void loadfloatOOS(String inputFile)

printStats

protected static void printStats(double[] ar)

castToDoubleArr

protected static double[] castToDoubleArr(float[] f)

castToDoubleArr

protected static double[] castToDoubleArr(short[] f)

castToFloatArr

protected static float[] castToFloatArr(double[] f)

castToFloatArr

protected static float[] castToFloatArr(short[] f)

standardNormalisation

public static float[] standardNormalisation(float[] data)

Normalises the data in the specified array to be in range [0,1], with 0 as the minimum, and 1 as the maximum. RETURNS THE SAME ARRAY OBJECT - i.e. changes are made in place.

Parameters:

data -

min

public static final float min(float[] a)

Return the min of the specified array

Parameters:

a - the array

Returns:

the minimum value in the arrays

max

public static final float max(float[] a)

Return the max of the specified array

Parameters:

a - the array

Returns:

the maximum value in the arrays