In this task you have to build a decision trees to be used as classifier, i.e., the aim of this task is to build a classifier using decision trees Dataset You will use the gene expression data (same one you have used for programming assignment #1) which contains 181 genes, each represented by 500 attributes. The last column of each sample represents the class for each gene. You should download this file and write a program (or programs) to do the following: What to do Choose attributes 50, 100, 250, and 500 (from now on named as A50, A100, A250, and A500). a) Find the best split value v for attribute A250 and discretize this attribute into two intervals such that one interval contains values of A250 V. V= ????? b) Discretize attributes A50, A100, A250, and A500 into 2 intervals using the equal frequency discretization method. c) Use the greedy approach discussed in Chapter 4 to build a 3-level decision tree using the attributes above. Use the Gini-index as the criterion for splitting using the best attribute (among the 5 considered). Make sure that the same attribute is not used more than once in any sub-tree. Discuss the tree computation process in your report giving data and reasons for choosing the attributes used for splitting at each level of the tree. Show the final tree obtained. A? Y