Decision Trees Quiz

A branching structure of sequential decisions

A flat list of independent rules

A hierarchical neural network

A linear regression model

By choosing the mean value of the feature

By evaluating all possible cutoffs between sorted values and selecting the one that maximizes information gain

By using the median value as the fixed threshold

By randomly selecting a value from the range of the feature

They assume a specific data distribution, making them highly specialized

They always provide the highest prediction accuracy compared to other models

They are simple, interpretable, and make no assumptions about data distribution

They require a very large amount of training data to work effectively

Growing a very deep tree and using all available data without any restrictions

Ignoring the impurity measures during node splitting

Increasing the number of features used at each split

Limiting the tree's depth and pruning back the tree based on performance on a validation set

0.42

0.48

0.54

0.60