ca.pfv.spmf.algorithms.sequentialpatterns.fournier2008_seqdim

Class AlgoPrefixSpanMDSPM

java.lang.Object

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

ca.pfv.spmf.algorithms.sequentialpatterns.fournier2008_seqdim.AlgoPrefixSpanMDSPM

public class AlgoPrefixSpanMDSPM
extends AbstractAlgoPrefixSpan

This is an implementation of the PrefixSpan algorithm by Pei et al. 2001 using optimizations such as pseudo-projections, designed specifically for being used with the multi-dimensional sequential pattern mining algorithm SeqDim. There is another implementation of PrefixSpan for general use that is more optimized and can be found in the package: ca.pfv.spmf.sequential_patterns/bide_and_prefixspan

See Also:

Sequence, SequenceDatabase, AlgoSeqDim

Constructor Summary

Constructors

Constructor and Description
AlgoPrefixSpanMDSPM(double minsup) Constructor

Method Summary

Modifier and Type	Method and Description
double	getMinSupp() Get the minsup threshold as a percentage (double).
void	printStatistics(int databaseSize) Print statistics about the latest execution of this algorithm.
Sequences	runAlgorithm(SequenceDatabase database) Run the algorithm

Methods inherited from class java.lang.Object

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

Constructor Detail

AlgoPrefixSpanMDSPM

public AlgoPrefixSpanMDSPM(double minsup)

Constructor

Parameters:

minsup - minimum support threshold as a percentage (double)

Method Detail

getMinSupp

public double getMinSupp()

Get the minsup threshold as a percentage (double).

Specified by:

getMinSupp in class AbstractAlgoPrefixSpan

Returns:

a double value

runAlgorithm

public Sequences runAlgorithm(SequenceDatabase database)

Run the algorithm

Specified by:

runAlgorithm in class AbstractAlgoPrefixSpan

Parameters:

database - a sequence database

Returns:

a set of sequential patterns (Sequences)

printStatistics

public void printStatistics(int databaseSize)

Print statistics about the latest execution of this algorithm.

Parameters:

databaseSize - the number of sequences in the original sequence database.