Use One-Hot Encoding for Categorical Features MACHINE-LEARNING-ASSOCIATE Practice Questions (Page 3)

Questions 11–15

1. 11foundation · easy

   What is the benefit of placing one-hot encoding inside a Spark ML pipeline?

   • AIt ensures the encoding is applied consistently to both training and inference data.
   • BIt reduces the number of features created by one-hot encoding.
   • CIt eliminates the need for a separate validation dataset.
   • DIt automatically selects the best encoding method for each column.
   Reveal Solution Discussion Select an answer first
2. 12application · medium

   A data engineer is using Spark ML to build a pipeline that one-hot encodes a categorical column 'payment_method' with values 'credit_card', 'debit_card', 'paypal'. They want to ensure that the OneHotEncoder stage is fitted only on the training data and then applied to test data without refitting. How should they set up the pipeline?

   • AFit the entire pipeline on the training data, then use the fitted pipeline to transform the test data.
   • BFit the OneHotEncoder on the test data first, then apply it to the training data.
   • CFit the StringIndexer on the training data, but fit the OneHotEncoder on the test data to adapt to new categories.
   • DFit the StringIndexer and OneHotEncoder on the combined training and test data, then train the model.
   Reveal Solution Discussion Select an answer first
4. 13application · easy

   A data analyst is preparing a dataset for a machine learning model. They have a categorical column 'color' with values 'red', 'green', 'blue'. They decide to use one-hot encoding. What is the primary purpose of this encoding?

   • ATo convert the categorical values into a numerical format that machine learning algorithms can process.
   • BTo sort the categories alphabetically for easier analysis.
   • CTo reduce the number of features in the dataset.
   • DTo replace missing values in the categorical column.
   Reveal Solution Discussion Select an answer first
5. 14foundation · easy

   What is the primary purpose of one-hot encoding for categorical variables?

   • ATo replace missing values in categorical columns with the most frequent category.
   • BTo convert each category into a binary column, where a 1 indicates the presence of that category and 0 indicates absence.
   • CTo assign a unique integer to each category, preserving any natural order among the categories.
   • DTo reduce the number of columns by grouping rare categories into a single 'other' category.
   Reveal Solution Discussion Select an answer first
6. 15application · medium

   A data scientist is evaluating the impact of one-hot encoding on a logistic regression model's interpretability. They have a categorical feature 'city' with 20 categories. What is the primary effect of one-hot encoding on model interpretability?

   • AIt combines all cities into a single coefficient, making interpretation simpler but less detailed.
   • BIt has no effect on interpretability because the model treats all features equally.
   • CIt makes the model completely uninterpretable because of the high dimensionality.
   • DIt creates a separate coefficient for each city, making it easy to understand the effect of each city on the prediction.
   Reveal Solution Discussion Select an answer first