org.conqat.lib.commons.datamining

Class FrequentItemSetMiner<T>

java.lang.Object

org.conqat.lib.commons.datamining.FrequentItemSetMiner<T>

public class FrequentItemSetMiner<T>
extends java.lang.Object

A-priori algorithm for mining frequent item sets from shopping baskets. See http://en.wikipedia.org/wiki/Apriori_algorithm

Constructor Summary

Constructors

Constructor and Description
FrequentItemSetMiner(double supportThreshold) Constructs a new FrequentItemSetMiner.

Method Summary

Modifier and Type	Method and Description
java.util.Set<FrequentItemSet<T>>	mineFrequentItemSets(java.util.Set<java.util.Set<T>> baskets) Mines frequent item sets from the given shopping baskets.

Methods inherited from class java.lang.Object

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

Constructor Detail

FrequentItemSetMiner

public FrequentItemSetMiner(double supportThreshold)

Constructs a new FrequentItemSetMiner.

Parameters:

supportThreshold - [0..1] denotes in what fraction of baskets an item set must occur to be considered frequent.

Method Detail

mineFrequentItemSets

public java.util.Set<FrequentItemSet<T>> mineFrequentItemSets(java.util.Set<java.util.Set<T>> baskets)

Mines frequent item sets from the given shopping baskets. The support threshold is the fraction of baskets from which a frequent item set is a subset. The commodity factor is used to ignore items that are purchased extremely often. If an item occurs in the fraction commodityFactor of the baskets, it is ignored for identifying frequent item sets. Elements in baskets must be hashable.

Parameters:

baskets - the baskets to be analyzed.