Primary Author is Sriram S (srirams{at}gmail{dotcom})

This code takes a list of bit vectors after features are extracted and performs association rule mining by the way of frequent item sets.

• Computes frequent item sets that are larger than some given threshold cutoff frequency.

• For each frequent item set with two or more item, it considers possible association rules derived from this set. For each association rule antecedent => consequence, it computes the ratio of frequency(antecedent /\ consequent)/frequency(antecedent).

All sources are in src/.java Compile them into corresponding class files bin/associationRuleMiner/.class

run

cd bin
java associationRuleMiner/LeMain ../clusteringBv.txt