1. [60 pts.] Apriori v.s. FP-Growth: Consider the following transaction database: ABCD, ACDF, ACDEG, ABDF, BCG, DFG, ABG, CDFG) (a) [20 pts.] Apriori: Show the execution of Apriori on this dataset. On each level you should clearly show the set of candidates and then the set of surviving frequent itemsets at this level. Use minsup=2 (b) [5 pts.] Effectiveness: How many more candidates would the Brute Force algorithm consider compared to those by Apriori (minsup-2)? (c) [25 pts.] FP-Tree: Create the FP-tree corresponding to the dataset by (1) filtering out infrequent singletons; (2) reordering the itemsets by decreasing singleton frequencies and (3) buildng the FP-tree by placing the reordered baskets one at a time in the= FP-tree. Use minsup=2. Show the resulting FP-tree (d) [10 pts.] FP-Growth: Compute the D-conditional pattern set for the tree from the previous part. Build the D-conditional FP-tree. 1. [60 pts.] Apriori v.s. FP-Growth: Consider the following transaction database: ABCD, ACDF, ACDEG, ABDF, BCG, DFG, ABG, CDFG) (a) [20 pts.] Apriori: Show the execution of Apriori on this dataset. On each level you should clearly show the set of candidates and then the set of surviving frequent itemsets at this level. Use minsup=2 (b) [5 pts.] Effectiveness: How many more candidates would the Brute Force algorithm consider compared to those by Apriori (minsup-2)? (c) [25 pts.] FP-Tree: Create the FP-tree corresponding to the dataset by (1) filtering out infrequent singletons; (2) reordering the itemsets by decreasing singleton frequencies and (3) buildng the FP-tree by placing the reordered baskets one at a time in the= FP-tree. Use minsup=2. Show the resulting FP-tree (d) [10 pts.] FP-Growth: Compute the D-conditional pattern set for the tree from the previous part. Build the D-conditional FP-tree