MS-DS Master of Data science Feature Engineering

Question 1

Which technique is used to reduce the impact of outliers when scaling numerical features?

Accepted Answer

Robust Scaling

Answer

Min-Max Normalization

Answer

Log Transformation

Answer

One-Hot Encoding

Question 2

What is the primary purpose of one-hot encoding in feature engineering?

Accepted Answer

Convert nominal categorical variables into binary indicator columns

Answer

Reduce dimensionality of numerical features

Answer

Convert ordinal categories into ranks

Answer

Normalize skewed distributions

Question 3

Label encoding is most appropriate for which type of categorical variable?

Accepted Answer

Ordinal variables with a meaningful rank order

Answer

Nominal variables with many unique values

Answer

Binary variables only

Answer

Continuous variables with missing values

Question 4

Which feature transformation is most commonly applied to correct a right-skewed (positively skewed) distribution?

Accepted Answer

Applying a log transformation

Answer

Squaring the feature

Answer

Applying min-max scaling

Answer

Applying principal component analysis

Question 5

What does 'feature interaction' refer to in the context of feature engineering?

Accepted Answer

Creating new features by combining two or more existing features

Answer

Removing correlated features from a dataset

Answer

Imputing missing values using surrounding feature values

Answer

Reducing the number of features via dimensionality reduction

Question 6

Which imputation strategy is most appropriate when data is missing completely at random (MCAR) and the dataset is small?

Accepted Answer

Multiple imputation using chained equations (MICE)

Answer

Dropping all rows with missing values

Answer

Replacing with the most frequent value

Answer

Replacing with a constant value of 0

Question 7

What is target encoding and what is its main risk?

Accepted Answer

Replacing a category with the mean of the target variable for that category; risk is data leakage

Answer

Encoding the target variable as a float; risk is data type mismatch

Answer

Encoding all features relative to the target range; risk is overfitting the target

Answer

Normalizing the target variable; risk is losing interpretability