AI Engineering Degree Practice Exam 2025 - Free AI Engineering Practice Questions and Study Guide

Question: 1 / 400

Which statement about decision trees is NOT true?

They can easily handle both numerical and categorical data.

They often require feature scaling to improve performance.

The statement that decision trees often require feature scaling to improve performance is not true. Decision trees inherently work by splitting the data based on the feature values at each node, evaluating one feature at a time for the best split. This means that they can naturally handle data without the need for feature scaling, such as normalization or standardization, which is often essential for algorithms that compute distances, like k-nearest neighbors or gradient descent-based methods.

Because decision trees make decisions based on the order and value of features rather than their specific scales, they maintain effective performance without transformed ranges. Thus, the requirement for feature scaling is not applicable to decision trees, setting this statement apart as incorrect in the context of their functionality.

Get further explanation with Examzify DeepDiveBeta

They work by dividing the dataset into subsets based on feature values.

They can overfit the training data if not properly constrained.

Next Question

Report this question

Download AI Engineering Degree Practice Exam 2025 - Free AI Engineering Practice Questions and Study Guide PDF Study Guide
Subscribe

Get the latest from Examzify

You can unsubscribe at any time. Read our privacy policy