Let I = { I1, I2, I3, , I10 } be a set of items, where Ij denotes an item ID. Consider the transaction database D, defined in the table below:

Transaction ID List of Items in the Transaction

T1 I1, I3, I4

T2 I1, I2

T3 I5, I6, I7,I9

T4 I1, I2, I6,I10

T5 I1, I6, I8,I9

(3) What is the occurrence frequency of itemsets

(a) { I1, I3, I4 }

(b) { I6, I9 }

(4) Suppose that the minimum support count threshold is set to 30%,

(a) is { I1, I3, I4 } a frequent itemset?

(b) is { I6, I9 } a frequent itemset?

(5) The Apriori algorithm is an approach for mining frequent itemsets and association rules.

(a) Use Apriori algorithm to find the frequent itemsets in the transaction database D. Assume min_sup = 2.

(b) Generate all strong association rules from one of the itemsets obtained in the step (a). Use minimum confidence threshold = 80%.