org.processmining.markovclustering.utils

Class MatrixUtils

java.lang.Object

org.processmining.markovclustering.utils.MatrixUtils

public class MatrixUtils
extends java.lang.Object

Constructor Summary

Constructors

Constructor and Description
MatrixUtils()

Method Summary

Modifier and Type	Method and Description
static void	columnNormalize(MCLMatrix matrix) Column-normalizes a Matrix to 1.
static double	getTotalAbsoluteChange(MCLMatrix matrixA, MCLMatrix matrixB) Get the total absolute change in values between two Matrices A and B.
static double	getTotalPercentageChange(MCLMatrix matrixA, MCLMatrix matrixB) Get the total percentage change in values between two Matrices A and B.
static void	prune(MCLMatrix matrix, double minThreshold) Prunes a matrix by setting the values of the matrix to 0 if they're lower than the minimal value threshold.

Methods inherited from class java.lang.Object

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

Constructor Detail

MatrixUtils

public MatrixUtils()

Method Detail

getTotalAbsoluteChange

public static double getTotalAbsoluteChange(MCLMatrix matrixA,
                                            MCLMatrix matrixB)

Get the total absolute change in values between two Matrices A and B.

Parameters:

matrixA - Matrix A

matrixB - Matrix B

Returns:

The total absolute change in values between matrices A and B

getTotalPercentageChange

public static double getTotalPercentageChange(MCLMatrix matrixA,
                                              MCLMatrix matrixB)

Get the total percentage change in values between two Matrices A and B.

Parameters:

matrixA - Matrix A

matrixB - Matrix B

Returns:

The total percentage change in values between matrices A and B

columnNormalize

public static void columnNormalize(MCLMatrix matrix)

Column-normalizes a Matrix to 1.

Parameters:

matrix - The matrix to normalize

prune

public static void prune(MCLMatrix matrix,
                         double minThreshold)

Prunes a matrix by setting the values of the matrix to 0 if they're lower than the minimal value threshold. This increases speed, because, in MCL, values that are low enough will never get big again in later stages.

Parameters:

matrix - The matrix to prune

minThreshold - The pruning threshold