ML2 PDFs Trees

What is the main characteristic of decision trees?

• They are very complex and difficult to understand

• They are not suitable for discovering hidden patterns

• They are easy to understand and implement

• They are only used for regression analysis

What is the measure of how impure or heterogeneous groups are after a split?

• Impurity

• Node

• Split value

• Loss

What is the maximum depth of a decision tree related to?

• The measure of how impure or heterogeneous groups are after a split

• The maximum length between the root node to a leaf

• The minimum number of observations that must be present in a partition to be split in a node

• The minimum number of observations that must be contained in a leaf

What is the purpose of pruning in decision trees?

• To increase overfitting

• To reduce the number of branches and overfitting

• To decrease the variance in the estimator

• To make the tree more complex

When is it good to use decision trees?

• When looking for predictions and explainability

• When the classes are imbalanced

• When a lot of data preprocessing is needed

• When looking for predictions but not explainability

What is the type of decision trees in terms of machine learning?

• For clustering

• Only for regression

• For both regression and classification

• Only for classification

What is the measure of how mixed the data is within a partition?

• Impurity

• Loss

• Split value

• Node

What is the main drawback of a decision tree if not balanced properly?

• Underfitting

• Low variance

• High bias

• Overfitting

What is the memory requirement for decision trees?

• Depends on the Data

• Large

• Medium

• Small

What is the training time for decision trees?

• Depends on the Data

• Fast

• Slow

• Instant