ca.pfv.spmf.algorithms.sequentialpatterns.spade_spam_AGP

Class AlgoSPADE

java.lang.Object

ca.pfv.spmf.algorithms.sequentialpatterns.spade_spam_AGP.AlgoSPADE

public class AlgoSPADE
extends java.lang.Object

This is an implementation of the SPADE. SPADE was proposed by ZAKI in 2001. NOTE: This implementation saves the pattern to a file as soon as they are found or can keep the pattern into memory, depending on what the user choose. Copyright Antonio Gomariz Peñalver 2013 This file is part of the SPMF DATA MINING SOFTWARE (http://www.philippe-fournier-viger.com/spmf). SPMF is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version. SPMF is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details. You should have received a copy of the GNU General Public License along with SPMF. If not, see .

Field Summary

Fields

Modifier and Type	Field and Description
long	end Start and end points in order to calculate the overall time taken by the algorithm
long	joinCount
long	start Start and end points in order to calculate the overall time taken by the algorithm

Constructor Summary

Constructors

Constructor and Description
AlgoSPADE(double support, boolean dfs, AbstractionCreator abstractionCreator) Constructor of the class that calls SPADE algorithm.

Method Summary

Modifier and Type	Method and Description
void	clear() It clears all the attributes of AlgoPrefixSpan class
double	getMinSupRelative() It gets the minimum relative support, i.e.
int	getNumberOfFrequentPatterns()
long	getRunningTime() It gets the time spent by the algoritm in its execution.
java.lang.String	printStatistics()
void	runAlgorithm(SequenceDatabase database, CandidateGenerator candidateGenerator, boolean keepPatterns, boolean verbose, java.lang.String outputFilePath) Actual call to SPADE algorithm.
void	runAlgorithmParallelized(SequenceDatabase database, CandidateGenerator candidateGenerator, boolean keepPatterns, boolean verbose, java.lang.String outputFilePath) Actual call to SPADE algorithm.

Methods inherited from class java.lang.Object

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

Field Detail

joinCount

public long joinCount

start

public long start

Start and end points in order to calculate the overall time taken by the algorithm

end

public long end

Start and end points in order to calculate the overall time taken by the algorithm

Constructor Detail

AlgoSPADE

public AlgoSPADE(double support,
                 boolean dfs,
                 AbstractionCreator abstractionCreator)

Constructor of the class that calls SPADE algorithm.

Parameters:

support - Minimum support (from 0 up to 1)

dfs - Flag for indicating if we want a depth first search. If false, we indicate that we want a breath-first search.

abstractionCreator - An abstraction creator.

Method Detail

runAlgorithm

public void runAlgorithm(SequenceDatabase database,
                         CandidateGenerator candidateGenerator,
                         boolean keepPatterns,
                         boolean verbose,
                         java.lang.String outputFilePath)
                  throws java.io.IOException

Actual call to SPADE algorithm. The output can be either kept or ignore. Whenever we choose to keep the patterns found, we can keep them in a file or in the main memory

Parameters:

database - Original database in where we want to search for the frequent patterns.

candidateGenerator - The candidate generator used by the algorithm SPADE

keepPatterns - Flag indicating if we want to keep the output or not

verbose - Flag for debugging purposes

outputFilePath - Path of the file in which we want to store the frequent patterns. If this value is null, we keep the patterns in the main memory. This argument is taken into account just when keepPatterns is activated.

Throws:

java.io.IOException

runAlgorithmParallelized

public void runAlgorithmParallelized(SequenceDatabase database,
                                     CandidateGenerator candidateGenerator,
                                     boolean keepPatterns,
                                     boolean verbose,
                                     java.lang.String outputFilePath)
                              throws java.io.IOException

Actual call to SPADE algorithm. The output can be either kept or ignore. Whenever we choose to keep the patterns found, we can keep them in a file or in the main memory. The algorithm SPADE is executed in a parallel way.

Parameters:

database - Original database in where we want to search for the frequent patterns.

candidateGenerator - The candidate generator used by the algorithm SPADE

keepPatterns - Flag indicating if we want to keep the output or not

verbose - Flag for debugging purposes

outputFilePath - Path of the file in which we want to store the frequent patterns. If this value is null, we keep the patterns in the main memory. This argument is taken into account just when keepPatterns is activated.

Throws:

java.io.IOException

printStatistics

public java.lang.String printStatistics()

getNumberOfFrequentPatterns

public int getNumberOfFrequentPatterns()

getRunningTime

public long getRunningTime()

It gets the time spent by the algoritm in its execution.

Returns:

getMinSupRelative

public double getMinSupRelative()

It gets the minimum relative support, i.e. the minimum number of database sequences where a pattern has to appear

Returns:

clear

public void clear()

It clears all the attributes of AlgoPrefixSpan class