MS-DS Master of Data science MS-DS Master of Data science Data Wrangling and Preprocessing Questions and Answers 3

Question 1

When merging two datasets with a many-to-many relationship on the join key, what is the most critical preprocessing step?

Accepted Answer

Deduplicating or reshaping one dataset to establish a clear one-to-many relationship

Answer

Sorting both datasets alphabetically

Answer

Converting all columns to string type

Answer

Applying PCA to reduce dimensionality before joining

Question 2

Which encoding strategy is most appropriate for a categorical variable with a natural ordering such as education level?

Accepted Answer

Ordinal encoding that preserves the rank order

Answer

One-hot encoding

Answer

Binary encoding

Answer

Frequency encoding

Question 3

What problem does target encoding introduce that requires careful mitigation through cross-validation or smoothing?

Accepted Answer

Data leakage from the target variable into the features

Answer

Loss of all categorical information

Answer

Increased dimensionality of the feature space

Answer

Automatic removal of rare categories

Question 4

During text data preprocessing for a data science pipeline, what is the correct order of operations?

Accepted Answer

Lowercasing, tokenization, stop word removal, stemming or lemmatization

Answer

Stemming, tokenization, uppercasing, stop word removal

Answer

Stop word removal, lemmatization, lowercasing, tokenization

Answer

Tokenization, stemming, uppercasing, stop word addition

Question 5

What is the primary risk of applying feature scaling before splitting data into training and test sets?

Accepted Answer

Information from the test set leaks into the training set through shared scaling parameters

Answer

The model will always underfit the training data

Answer

Categorical variables will be incorrectly converted

Answer

The dataset will lose all missing values

Question 6

Which resampling technique for handling class imbalance generates synthetic minority samples by interpolating between existing minority observations?

Accepted Answer

SMOTE (Synthetic Minority Over-sampling Technique)

Answer

Random undersampling

Answer

Stratified k-fold cross-validation

Answer

Bootstrap aggregation