ca.pfv.spmf.algorithms.frequentpatterns.vme

Class AlgoVME

java.lang.Object

ca.pfv.spmf.algorithms.frequentpatterns.vme.AlgoVME

public class AlgoVME
extends java.lang.Object

This is an implementation of the VME algorithm (Deng and Xu, 2011) for erasable itemset mining.

The VME algorithm finds all the ereasable itemsets from a product database.

Actually, this algorithms is a only slight modification of the AprioriTID algorithm.

I have implemented mostly as described in the paper with some modifications to make it more efficient.
First, the authors suggested to generate all candidates of a level before removing the unereasable ones. This is inefficient. Instead, in my implementation, I check the "gain" (loss of profit) directly after generating a candidate so I can eliminate them right away.
Second, it is unecessary to check the subsets like the authors suggest because they use a vertical representation.
Third, the authors suggest to store the profit of transactions in PID List. This is not memory efficient. For implementation it is better to store the profit of each transaction only once in a hashtable.

See Also:

Itemset

Constructor Summary

Constructors

Constructor and Description
AlgoVME() Default constructor

Method Summary

Modifier and Type	Method and Description
void	printStats() Print statistics about the latest execution to System.out.
void	runAlgorithm(java.lang.String input, java.lang.String output, double threshold) Run the VME algorithm.

Methods inherited from class java.lang.Object

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

Constructor Detail

AlgoVME

public AlgoVME()

Default constructor

Method Detail

runAlgorithm

public void runAlgorithm(java.lang.String input,
                         java.lang.String output,
                         double threshold)
                  throws java.lang.NumberFormatException,
                         java.io.IOException

Run the VME algorithm.

Parameters:

input - path to an input file

output - path to be used for writing the output file

threshold - the threshold chosen by the user.

Throws:

java.io.IOException - exception if error reading/writing files

java.lang.NumberFormatException

printStats

public void printStats()

Print statistics about the latest execution to System.out.