Data Science Data Wrangling and Preprocessing 3

Question 1

A feature ranges from 0 to 1,000,000 while another ranges from 0 to 1. Which preprocessing step puts them on a comparable scale?

Accepted Answer

Feature scaling (normalization/standardization)

Answer

One-hot encoding

Answer

Deduplication

Answer

Tokenization

Question 2

Min-max scaling transforms a feature to which range by default?

Accepted Answer

[0, 1]

Answer

[-1, 1]

Answer

Mean 0, std 1

Answer

[0, 100]

Question 3

Standardization (z-score) rescales a feature to have what properties?

Accepted Answer

Mean 0 and standard deviation 1

Answer

Min 0 and max 1

Answer

Sum equal to 1

Answer

Median 0 and IQR 1

Question 4

When encoding a nominal categorical variable with no inherent order, which method avoids implying a false ranking?

Accepted Answer

One-hot encoding

Answer

Label/ordinal encoding

Answer

Min-max scaling

Answer

Log transform

Question 5

A right-skewed feature like website session duration is transformed with a log. What is the main benefit?

Accepted Answer

It reduces skewness and compresses large values

Answer

It removes all missing data

Answer

It converts it to categorical

Answer

It guarantees a uniform distribution

Question 6

Why should scaling parameters be computed only on the training set, then applied to the test set?

Accepted Answer

To prevent data leakage from the test set

Answer

To make the test set larger

Answer

To remove duplicates

Answer

To convert types

Question 7

Which scaling method is most robust when a feature contains many extreme outliers?

Accepted Answer

RobustScaler (uses median and IQR)

Answer

Min-max scaling

Answer

Standard scaling

Answer

Unit-vector scaling