ca.pfv.spmf.algorithms.frequentpatterns.hmine

Class AlgoHMine

java.lang.Object

ca.pfv.spmf.algorithms.frequentpatterns.hmine.AlgoHMine

public class AlgoHMine
extends java.lang.Object

An implementation of the HMine algorithm for mining frequent itemsets from a transaction database.

It is based on the description in:

Pei et al. (2007) H-Mine: Fast and space-preserving frequent pattern mining in large databases. IIE Transactions, 39, 593-605.

I tried to follow as much as possible the description in the article for HMine(mem). One observation is that the links for an item in the header table are simply what is called a "tid set" in some other algorithms, because links always point to the first element of a transaction. So actually, the algorithm was more simple than I first thought.

Constructor Summary

Constructors

Constructor and Description
AlgoHMine() Default constructor

Method Summary

Modifier and Type	Method and Description
void	printStatistics() Print statistics about the latest execution of the algorithm to System.out.
void	runAlgorithm(java.lang.String input, java.lang.String output, int minsup) Run the algorithm.

Methods inherited from class java.lang.Object

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

Constructor Detail

AlgoHMine

public AlgoHMine()

Default constructor

Method Detail

runAlgorithm

public void runAlgorithm(java.lang.String input,
                         java.lang.String output,
                         int minsup)
                  throws java.io.IOException

Run the algorithm.

Parameters:

input - the path of the input file (a transaction database)

output - the output file path for writing the result

minsup - the minimum support threshold

Throws:

java.io.IOException - exception if error while writing the file

printStatistics

public void printStatistics()

Print statistics about the latest execution of the algorithm to System.out.