1. Which split value for age would best separate the newspaper subscribers from nonsubscribers based on the Gini impurity index?

2. Which split value for income would best separate the newspaper subscribers from nonsubscribers based on the Gini impurity index?

3. Between the best split values for age and income, which one should be chosen as the first split for a decision tree that predicts whether someone is a newspaper subscriber or not?

4. State the rules generated from this split.

A B C D E 1 Age Income Subscribe 2 25 43000 0 37 42000 1 45 86000 0 29 36500 6 57 112000 1 43 54000 8 39 68000 LO 43 42000 10 11 12 13 14