Consider the training examples shown in the following table for a binary classification problem. CID | Gender Type 1 Size 1 Class 1 CO | 1 | 2 3 4 | I 1 CO CO CO I 1 5 6 7 8 9 CO CO CO 1 I 1 CO CO | 1 I | M M M M M M F F F F M M M M F F F F F F Family Sports Sports Sports Sports Sports Sports Sports Sports Luxury Family Family Family Luxury Luxury Luxury Luxury Luxury Luxury Luxury 10 11 12 13 14 15 16 17 18 19 20 | Small 1 | Medium 1 | Medium 1 | Large | E Large 1 | E Large 1 | Small 1 | Small | Medium | Large 1 | Large 1 | E Large | Medium | E Large | Small | Small | | Medium | Medium | | Medium | | Large 1 CO C1 | | 1 I 1 | | I C1 C1 C1 C1 C1 C1 C1 C1 C1 | (d) What is the information gain if the split is based on Size? | (e) Which attribute provides the best split if information gain is the splitting criterion. (f) Based on the given training date set, calculate the probabilities required for a Nave Bayers classifier. Using Laplace smoothing for k=1 estimate to smooth the following probability estimates: P(C=CO) = ? P(C=C1) = ? P(G=M|C=C1)= ? P(G=F|C=CO) = ? P(G=M|C=CO) = ? P(G=F|C=C1) = ? P(T=F|C=CO) = ? P(T =F|C=C1) = ? P(T =S|C=CO) = ? P(T =S|C=C1) = ? P(T =L|C=C0) = ? P(T =L|C=C1) = ? P(S=S|C=CO) = ? P(S=S|C=C1)= ? P(S=MC=CO) = ? P(S=M|C=C1) = ? P(S=L|C=CO) = ? P(S=LC=C1) = ? P(S=EC=CO) = ? P(S=EC=C1) = ? Consider the training examples shown in the following table for a binary classification problem. CID | Gender Type 1 Size 1 Class 1 CO | 1 | 2 3 4 | I 1 CO CO CO I 1 5 6 7 8 9 CO CO CO 1 I 1 CO CO | 1 I | M M M M M M F F F F M M M M F F F F F F Family Sports Sports Sports Sports Sports Sports Sports Sports Luxury Family Family Family Luxury Luxury Luxury Luxury Luxury Luxury Luxury 10 11 12 13 14 15 16 17 18 19 20 | Small 1 | Medium 1 | Medium 1 | Large | E Large 1 | E Large 1 | Small 1 | Small | Medium | Large 1 | Large 1 | E Large | Medium | E Large | Small | Small | | Medium | Medium | | Medium | | Large 1 CO C1 | | 1 I 1 | | I C1 C1 C1 C1 C1 C1 C1 C1 C1 | (d) What is the information gain if the split is based on Size? | (e) Which attribute provides the best split if information gain is the splitting criterion. (f) Based on the given training date set, calculate the probabilities required for a Nave Bayers classifier. Using Laplace smoothing for k=1 estimate to smooth the following probability estimates: P(C=CO) = ? P(C=C1) = ? P(G=M|C=C1)= ? P(G=F|C=CO) = ? P(G=M|C=CO) = ? P(G=F|C=C1) = ? P(T=F|C=CO) = ? P(T =F|C=C1) = ? P(T =S|C=CO) = ? P(T =S|C=C1) = ? P(T =L|C=C0) = ? P(T =L|C=C1) = ? P(S=S|C=CO) = ? P(S=S|C=C1)= ? P(S=MC=CO) = ? P(S=M|C=C1) = ? P(S=L|C=CO) = ? P(S=LC=C1) = ? P(S=EC=CO) = ? P(S=EC=C1) =