Data Science FREE Data Science Data Wrangling and Preprocessing Questions and Answers 3

Question 1

What is the primary risk of using Min-Max scaling on a dataset before splitting into training and test sets?

Accepted Answer

It introduces data leakage because test set statistics influence the transformation

Answer

It increases computational complexity exponentially

Answer

It removes all outliers from the dataset

Answer

It converts continuous variables into categorical ones

Question 2

Which pandas method is most efficient for converting a wide-format DataFrame with repeated measurements into long format?

Accepted Answer

melt()

Answer

pivot_table()

Answer

groupby()

Answer

merge()

Question 3

When dealing with a heavily right-skewed target variable in a regression task, which preprocessing step is most commonly applied?

Accepted Answer

Box-Cox or log transformation

Answer

Standardization using z-scores

Answer

Removing the top 50% of values

Answer

Binning the variable into equal-width categories

Question 4

What is the purpose of using stratified sampling when splitting a dataset into training and test sets?

Accepted Answer

To maintain the same class distribution in both the training and test sets

Answer

To ensure both sets have equal numbers of features

Answer

To randomly shuffle the data before splitting

Answer

To remove imbalanced classes from the dataset

Question 5

Which approach is recommended for handling high-cardinality categorical features with hundreds of unique values?

Accepted Answer

Target encoding or frequency encoding

Answer

One-hot encoding all values

Answer

Dropping the feature entirely

Answer

Converting to a single binary column

Question 6

What does the Interquartile Range (IQR) method define as an outlier?

Accepted Answer

Any value below Q1 minus 1.5 times IQR or above Q3 plus 1.5 times IQR

Answer

Any value more than 2 standard deviations from the mean

Answer

Any value that appears fewer than 5 times in the dataset

Answer

Any value in the bottom or top 5% of the distribution