ca.pfv.spmf.algorithms.sequentialpatterns.fournier2008_seqdim

Class AlgoFournierViger08

java.lang.Object

ca.pfv.spmf.algorithms.sequentialpatterns.fournier2008_seqdim.AbstractAlgoPrefixSpan

ca.pfv.spmf.algorithms.sequentialpatterns.fournier2008_seqdim.AlgoFournierViger08

public class AlgoFournierViger08
extends AbstractAlgoPrefixSpan

This is the original implementation of the Fournier-Viger algorithm (2008) for sequential pattern mining, which combines features from several algorithms and includes original features such as accepting items with double values. For details about this algorithm see:

Fournier-Viger, P., Nkambou, R & Mephu Nguifo, E. (2008), A Knowledge Discovery Framework for Learning Task Models from User Interactions in Intelligent Tutoring Systems. Proceedings of the 7th Mexican International Conference on Artificial Intelligence (MICAI 2008). LNAI 5317, Springer, pp. 765-778.

This implementation can keep the result into memory and return it by the method runAlgorithm() or save the result directly to a file, if an output file path is provided.

See Also:

SequenceDatabase, Sequence, Sequences, PseudoSequence, PseudoSequenceDatabase, Pair, AlgoKMeansWithSupport, AbstractAlgoPrefixSpan

Constructor Summary

Constructors

Constructor and Description
AlgoFournierViger08(double minsupp, double minInterval, double maxInterval, double minWholeInterval, double maxWholeInterval, AlgoKMeansWithSupport algoClustering, boolean findClosedPatterns, boolean enableBackscanPruning)

Method Summary

Modifier and Type	Method and Description
double	getMinSupp() Get the minsup threshold as a percentage (doule)
int	getMinsuppRelative() Get the minsup threshold as an integer (sequence count)
void	printResult(int databaseSize) Print the seq.
void	printStatistics() Print statistics about the algorithm execution to System.out.
Sequences	runAlgorithm(SequenceDatabase database) Run the algorithm and save the result to memory
void	runAlgorithm(SequenceDatabase database, java.lang.String outputFilePath) Run the algorithm and save the result to a file

Methods inherited from class java.lang.Object

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

Constructor Detail

AlgoFournierViger08

public AlgoFournierViger08(double minsupp,
                           double minInterval,
                           double maxInterval,
                           double minWholeInterval,
                           double maxWholeInterval,
                           AlgoKMeansWithSupport algoClustering,
                           boolean findClosedPatterns,
                           boolean enableBackscanPruning)

Parameters:

minsupp - minimum support

minInterval - minimum item interval between two adjacent items. (C1)

maxInterval - maximum item interval between two adjacent items. (C2)

minWholeInterval - minimum item interval between the head and tail of a sequence. (C3)

maxWholeInterval - maximum item interval between the head and tail of a sequence (C4)

algoClustering - algorithm for clustering

findClosedPatterns - to mine only closed sequences

Method Detail

runAlgorithm

public void runAlgorithm(SequenceDatabase database,
                         java.lang.String outputFilePath)
                  throws java.io.IOException

Run the algorithm and save the result to a file

Parameters:

database - a sequence database

outputFilePath - an output file

Throws:

java.io.IOException - throw exception if error creating output file

runAlgorithm

public Sequences runAlgorithm(SequenceDatabase database)
                       throws java.io.IOException

Run the algorithm and save the result to memory

Specified by:

runAlgorithm in class AbstractAlgoPrefixSpan

Parameters:

database - a sequence database

Returns:

a set of sequential patterns (Sequences)

Throws:

java.io.IOException

printStatistics

public void printStatistics()

Print statistics about the algorithm execution to System.out.

printResult

public void printResult(int databaseSize)

Print the seq. patterns found to System.out. with

Parameters:

databaseSize - the size of the database (a number of sequences)

getMinSupp

public double getMinSupp()

Get the minsup threshold as a percentage (doule)

Specified by:

getMinSupp in class AbstractAlgoPrefixSpan

Returns:

a double

getMinsuppRelative

public int getMinsuppRelative()

Get the minsup threshold as an integer (sequence count)

Returns:

an integer