1. Consider the training data shown in the following table. (1) Compute the Gini index for the overall dataset. (8 points) (2) Compute the Gini index for the Customer ID attribute. (8 points) (3) Compute the Gini index for the Gender attribute. (8 points) (4) Compute the Gini index for the Car Type attribute using multiway split (3 splits). (8 points) (5) Compute the Gini index for the Shirt Size attribute using binary split (small/medium as one split and large/extra large as one split). (8 points) (6) Which attribute is better among Gender, Car Type, and Shirt Size? (8 points) (7) Why Customer ID is not a good attribute for splitting? (8 points) Training Data for Exercise 1 \begin{tabular}{|c|c|c|c|c|} \hline Customer ID & Gender & Car Type & Shirt Size & Class \\ \hline 1 & Male & Sports & Medium & C0 \\ \hline 2 & Male & Family & Small & C0 \\ \hline 3 & Male & Sports & Large & C0 \\ \hline 4 & Male & Sports & Medium & C0 \\ \hline 5 & Male & Sports & Extra Large & C0 \\ \hline 6 & Male & Sports & Extra Large & C0 \\ \hline 7 & Female & Sports & Small & C0 \\ \hline 8 & Female & Sports & Small & C0 \\ \hline 9 & Female & Sports & Medium & C0 \\ \hline 10 & Female & Luxury & Large & C0 \\ \hline 11 & Male & Family & Large & C1 \\ \hline 12 & Male & Family & Extra Large & C1 \\ \hline 13 & Male & Family & Medium & C1 \\ \hline 14 & Male & Luxury & Extra Large & C1 \\ \hline 15 & Female & Luxury & Small & C1 \\ \hline 16 & Female & Luxury & Small & C1 \\ \hline 17 & Female & Luxury & Medium & C1 \\ \hline 18 & Female & Luxury & Medium & C1 \\ \hline 19 & Female & Luxury & Medium & C1 \\ \hline 20 & Female & Luxury & Large & C1 \\ \hline \end{tabular}