Q

Question

A Decision Tree is a flowchart-like structure in which each internal node represents a "test" on an attribute (e.g., whether a coin flip comes up heads or tails), each branch represents the outcome of the test, and each leaf node represents a class label (a decision taken after computing all attributes). The paths from root to leaf represent classification rules.

The objective of a Decision Tree is to make decisions based on the attributes that best split the data into homogeneous sets (i.e., sets that mostly contain samples from the same class). This is because the final goal is to accurately classify new, unseen data.

The Gini impurity is a measure of how often a randomly chosen element from the set would be incorrectly labeled if it were randomly classified according to the distribution of labels in the set. The Gini impurity ranges between 0 and 1, where 0 implies the set is pure (all samples belong to the same class), and 1 implies the set is maximally impure (samples are distributed equally among all classes).

A. Maximise the Gini impurity of the leaf nodes - Incorrect, as we want to minimize the impurity. B. Minimise the homogeneity of the leaf nodes - Incorrect, as we want to maximize homogeneity (minimize impurity). C. Maximise the heterogeneity of the leaf nodes - Incorrect, as we want to minimize heterogeneity (maximize homogeneity). D. Minimise the Gini impurity of the leaf nodes - Correct, as we aim to create the purest possible sets at the leaf nodes.