Given a frequent itemset L, the goal when generating rules is to find all non-
empty subsets that satisfy the minimum confidence requirement. If |L| = k, then
there are 2k2 candidate association rules. So, as in the frequent itemset generation,
we need to find ways to generate rules efficiently. For the Apriori Algorithm we can
generate candidate rules by merging two rules that share the same prefix in the rule
consequent.