Data Scaling

Test your knowledge of normalization, standardization, and why data scaling is crucial for machine learning.

FREE QUIZ

8 Questions

~16 Minutes

Intermediate

Question 1

What is the primary purpose of data scaling in machine learning?

To increase model complexity To ensure consistent feature influence and avoid bias To convert data into binary format To eliminate all outliers

Question 2

Which of the following are two common methods of data scaling?

One-hot encoding and PCA Normalization and standardization Regression and classification Bagging and boosting

Question 3

Normalization is also referred to as:

Z-score transformation Min-max scaling Feature hashing Variance reduction

Question 4

Which value range is typically used in normalization?

[-100, 100] [-1, 1] or [0, 1] [mean ± std] Any range that includes outliers

Question 5

Standardization transforms data to have:

Values between 0 and 1 A mean of 1 and std of 10 A mean of 0 and standard deviation of 1 Discrete integer categories

Question 6

Why is data scaling important for optimization algorithms in neural networks?

It helps models overfit less It improves gradient convergence and learning speed It reduces memory usage It changes labels into numbers

Question 7

When should normalization be preferred over standardization?

When the data follows a normal distribution When features are skewed and need to fit a fixed scale When data is already centered When outliers must be retained

Question 8

What effect does unscaled data have on machine learning models?

It improves model generalization It has no effect if the dataset is large It may lead to biased predictions due to dominant features It always increases accuracy

Back to Quizzes

0 / 8 Answered | 0:00