•   

Table 1 shows a transaction database, D = {TID1, TID2, ..., TID7} at a store purchased by customers on five different skincare routine: Cleanser, Toner, Moisturizer, Serum, and Sunscreen. Let minimum support count = 2 and minimum confidence = 80%. Generate ALL frequent itemsets using Frequent Pattern (FP) Growth algorithm. Table 1. Transaction ID List of items in the transaction TID1 Moisturizer, Cleanser, Serum TID2 Cleanser, Toner TID3 Cleanser, Sunscreen TID4 Moisturizer, Cleanser, Toner TID5 Moisturizer, Sunscreen TID6 Cleanser, Sunscreen TID7 Moisturizer, Sunscreen TID8 Moisturizer, Cleanser, Sunscreen, Serum TID9 Moisturizer, Cleanser, Sunscreen a) Find the support count for each item, then sort the item list in descending order for their support count in Table 2. (4 marks) b) Construct FP tree for the transaction. Show and label all the nodes. (6 marks) c) Find the frequent patterns for item serum OR sunscreen. (6 marks) d) Generate TWO (2) rules for item serum OR sunscreen. Provide the confidence value for each rule. (4 marks)