Based on the following database of 5 transactions, use Frequent Pattern Growth algorithm with minimum support 0.4 to find frequent patterns TransactionID Items b,d,f,g,l f.g,h.l.m.n b,f,h.k.m a,f,h.j,m d,f,g.j,m 4 1. Generate an ordered list of frequent items based on the raw transaction database. Note that by items we mean length 1 patterns 2. Generate Header Table and FP-tree based on the frequent item list. Link nodes to the corresponding positions in the Header Table. 3. Generate Conditional Pattern Bases and Conditional FP-trees for items m,h,b.j based on the FP-tree, and list the frequent patterns computed based on each of the Condi- tional FP-trees. 4. Why do we order the items in each transaction by their frequency before constructing the FP-tree? 5. Which of the patterns generated in Part 3 are closed? Which are maximal? 6. List at least 2 association rules with minimum confidence- 0.6 from the frequent patterns computed in earlier parts Based on the following database of 5 transactions, use Frequent Pattern Growth algorithm with minimum support 0.4 to find frequent patterns TransactionID Items b,d,f,g,l f.g,h.l.m.n b,f,h.k.m a,f,h.j,m d,f,g.j,m 4 1. Generate an ordered list of frequent items based on the raw transaction database. Note that by items we mean length 1 patterns 2. Generate Header Table and FP-tree based on the frequent item list. Link nodes to the corresponding positions in the Header Table. 3. Generate Conditional Pattern Bases and Conditional FP-trees for items m,h,b.j based on the FP-tree, and list the frequent patterns computed based on each of the Condi- tional FP-trees. 4. Why do we order the items in each transaction by their frequency before constructing the FP-tree? 5. Which of the patterns generated in Part 3 are closed? Which are maximal? 6. List at least 2 association rules with minimum confidence- 0.6 from the frequent patterns computed in earlier parts