edu.ucla.sspace.matrix

Class CorrelationTransform.CorrelationGlobalTransform

java.lang.Object

edu.ucla.sspace.matrix.CorrelationTransform.CorrelationGlobalTransform

All Implemented Interfaces:

GlobalTransform

Enclosing class:

CorrelationTransform

public class CorrelationTransform.CorrelationGlobalTransform
extends Object
implements GlobalTransform

Constructor Summary

Constructors

Constructor and Description
CorrelationTransform.CorrelationGlobalTransform(File inputMatrixFile, MatrixIO.Format format) Creates an instance of CorrelationTransform from a File for format MatrixIO.Format.
CorrelationTransform.CorrelationGlobalTransform(Matrix matrix) Creates an instance of CorrelationTransform from a Matrix.

Method Summary

Methods

Modifier and Type	Method and Description
double	transform(int row, int column, double value) Computes the correlation, scaled using the square root, between item row and feature column where value specifies the number of occurances.

Methods inherited from class java.lang.Object

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

Constructor Detail

CorrelationTransform.CorrelationGlobalTransform

public CorrelationTransform.CorrelationGlobalTransform(Matrix matrix)

Creates an instance of CorrelationTransform from a Matrix.

CorrelationTransform.CorrelationGlobalTransform

public CorrelationTransform.CorrelationGlobalTransform(File inputMatrixFile,
                                               MatrixIO.Format format)

Creates an instance of CorrelationTransform from a File for format MatrixIO.Format.

Method Detail

transform

public double transform(int row,
               int column,
               double value)

Computes the correlation, scaled using the square root, between item row and feature column where value specifies the number of occurances. If value is zero, the correlation is zero.

Specified by:

transform in interface GlobalTransform

Parameters:

row - The index specifying the item being observed

column - The index specifying the feature being observed

value - The number of occurance of the item and feature

Returns:

the square root of the correlation between the item aand feature