edu.northwestern.at.morphadorner.corpuslinguistics.statistics

Class Collocation

java.lang.Object

edu.northwestern.at.morphadorner.corpuslinguistics.statistics.Collocation

public class Collocation
extends java.lang.Object

Computes bigram collocation measures.

Field Summary

Fields

Modifier and Type	Field and Description
static int	DICE Indices of association measures in result array.
static int	LOGLIKE
static int	PHISQUARED
static int	SCP
static int	SMI
static int	T
static int	Z

Constructor Summary

Constructors

Modifier	Constructor and Description
protected	Collocation() Don't allow instantiation but do allow overrides.

Method Summary

Methods

Modifier and Type	Method and Description
static double[]	association(int sampleCount, int refCount, int sampleSize, int refSize) Computes collocation measures.

Methods inherited from class java.lang.Object

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

Field Detail

DICE

public static final int DICE

Indices of association measures in result array.

See Also:

Constant Field Values

LOGLIKE

public static final int LOGLIKE

See Also:

Constant Field Values

PHISQUARED

public static final int PHISQUARED

See Also:

Constant Field Values

SMI

public static final int SMI

See Also:

Constant Field Values

SCP

public static final int SCP

See Also:

Constant Field Values

T

public static final int T

See Also:

Constant Field Values

Z

public static final int Z

See Also:

Constant Field Values

Constructor Detail

Collocation

protected Collocation()

Don't allow instantiation but do allow overrides.

Method Detail

association

public static double[] association(int sampleCount,
                   int refCount,
                   int sampleSize,
                   int refSize)

Computes collocation measures.

Parameters:

sampleCount - Count of collocation appearance in sample.

refCount - Count of collocation appearance in reference corpus.

sampleSize - Number of words/lemmas in the sample.

refSize - Number of words/lemmas in reference corpus.

Returns:

A double array containing the following measures of collocational association. (0) Dice coefficient (1) Log likelihood (2) Phi squared (3) Specific Mutual information score (4) Symmetric conditional probability (5) z score (6) t score