Please show your work. 1. Construct a binary DT for the following dataset. Please note each row in the table does NOT represent a single record. There are (5+ 20 + 25) positive and (20+5+25) negative in the data set (total 100 records) Number of instances Class-Class A B C TITIT F?T?T 20 20 T?T?F F?T?F T?F 25 25 1) Develop a DT using entropy measures. Split the DT until level 2 (level 1- root) so that the (deepest) leaf nodes are at level 3. After training the tree, calculate the total misclassification error rate (using the same dataset) Repeat question 1), but begin with C as the splitting criterion at the root. Compare misclassification rate of the two DTs. Which one is better? Based on the comparison, comment on the greedy strategy used in the DT algorithm. 2) 3) 2. Apply sequential covering algorithm to the same dataset to build the rule(s) for+ class