If a data analyst finds that a decision tree model has too many nodes or leaves to be meaningful, the analyst should apply _________ to the tree.

• A. pruning
• B. expansion
• C. trimming
• D. generalization

Answer: (A)

When a decision tree becomes too large with many nodes or leaves, the aim is to reduce its complexity so it generalizes better to new data. The technique that does this is pruning. Pruning removes branches that don’t add meaningful predictive power, typically by evaluating performance on separate validation data or using a criterion that penalizes complexity. This lowers variance and helps prevent overfitting, producing a simpler, more robust model.

Expansion would add more branches and worsen overfitting. Trimming isn’t the standard, widely used term for this specific process in decision trees. Generalization describes the desired outcome, not the specific operation to apply to the tree.