Frequent itemsets do no mean association rules. One more step is required to convert these frequent itemsets into rules.
Association Rules can be found from every frequent itemset X as follows:
For every non-empty subset A of X
Let B = X - A.
A Þ B is an association rule if
confidence(A Þ B) ≥ minConf.
where, confidence (A Þ B) = support (AB) / support (A), and
support(A Þ B) = support(AB).
Suppose X = 234 is a frequent itemset, with minSupp = 50%.
Proper non-empty subsets of X are: 23, 24, 34, 2, 3, 4 with supports = 50%, 50%, 75%, 75%, 75%, and 75%, respectively.
The association rules from these subsets are:
23 Þ 4. confidence = 100%.
24 Þ 3. confidence = 100%.
34 Þ 2. confidence = 67%.
2 Þ 34. confidence = 67%.
3 Þ 24. confidence = 67%.
4 Þ 23. confidence = 67%.
All rules have a support = 50%.
In order to derive an association rule A Þ B, we need to have support(AB) and support(A). This step is not as time consuming as the frequent itemset generation. It can also be speeded by using parallel processing techniques, as rules generated from one frequent itemset do not affect the rules generated from any other frequent itemset.
