Ivy_Honeywell_Data Analytics_7.2

A decision tree is built in __________ model

• Top-Down

• Bottom-Up

• All of the above

The goal of the feature selection while building a decision tree is to find features or attributes (decision nodes) which lead to split to children nodes which has ____________ entropy.

• Maximum

• Minimum

A data segment can be said to be pure if all the data instances belong to the _____________

• Single Class

• Multiple Class

The goal of feature selection while working with decision tree is to find the feature that results in _______________

• Maximum

• Least

Entropy value of ________________ represents that data sample has 50-50 split belonging to two categories

• 1

• 0

Entropy value of _______________ represents that data sample is pure or homogeneous

• 1

• 0

A decision tree created with the aim to achieve very minimal entropy may result in __________

• Overfitting

• Underfitting

Pre-pruning the decision tree may result in ______________

• Overfitting

• Underfitting

Which of the following is preferred strategy for building the most optimal decision tree?

• Pre-pruning

• Post-pruning

Which of the following example are not an example of ensemble learning algorithm?

• Random Forest

• Ada Boost

• Extra Trees

• Decision Trees