Decision Trees

A decision tree is an unsupervised machine learning approach.

• TRUE

• FALSE

What could happen to a decision tree if you don't properly manage its size?

• It could overfit.

• It could underfit.

• It could grow forever.

• It could become too robust.

What type of decision tree should you build in order to predict the salary of a worker based on their age and work experience?

• prediction tree

• classification tree

• regression tree

• supervised tree

Decision trees are built using a process that involves repeatedly splitting data into smaller and more homogenous subsets. What is this process called?

• information gain

• similarity

• homogeneity

• recursive partitioning

A partition with high SSR (Sum of Squared Residuals) is a partition with low variability.

• TRUE

• FALSE

If you allow a decision tree to grow as large as it can, and then reduce its size afterwards, what are you doing?

• pre-pruning

• downsizing

• post-pruning

• underfitting

The reduction in entropy that occurs as a result of a split is known as ________.

• information gain

• gain index

• Gini

• combined entropy

How is information gain used in decision trees to make splitting decisions?

• Information gain is used to randomly split the data in decision trees

• Information gain is used to determine the depth of the decision tree

• Information gain is used to calculate the accuracy of the decision tree

• Information gain is used to select the best attribute for splitting the data in decision trees.

What are the benefits of using pruning techniques in decision trees?

• Increasing overfitting

• Decreasing generalization

• Adding complexity

• The benefits of using pruning techniques in decision trees include preventing overfitting, improving generalization, and reducing complexity.

Calculate the entropy for a dataset with two classes, where one class has 60 instances and the other has 40 instances.

• 1.102

• 0.971

• 0.845

• 0.693