Now we need to find all the situations where customers who bought a subset of a frequent
itemset, most of the time also bought the remaining items in the same frequent itemset. Given a
frequent itemset, say (S1, S2, S3), if a customer who buys a subsert formed by S1 and S2, also
buys S3 80% of the times, then it is worth to consider the rule. This percentage is called the
confidence of the rule and is defined as the ratio of the number of transactions that include all
items in a particular frequent itemset to the number of transactions that include all items in the
subset.