Virtual Labs

Data Preprocessing and Feature Engineering

How is the Cabin attribute handled during preprocessing in this experiment?

a: It is encoded using One-Hot Encoding after preprocessing Explanation

Explanation

b: It is removed from the dataset due to missing values Explanation

Explanation

c: It is converted into a numerical attribute using Label Encoding Explanation

Explanation

d: It is normalized using feature scaling techniques Explanation

Explanation

Suppose a numerical attribute contains several extreme outliers. Which imputation method would generally be more appropriate than mean imputation?

a: Median imputation Explanation

Explanation

b: Mode imputation Explanation

Explanation

c: Label encoding Explanation

Explanation

d: One-hot encoding Explanation

Explanation

Why was the mode (most frequent value) used to fill missing values in the Embarked attribute?

a: Because categorical variables cannot be averaged Explanation

Explanation

b: Because it increases the dataset size Explanation

Explanation

c: Because it removes duplicate values Explanation

Explanation

d: Because it reduces dimensionality Explanation

Explanation

Why must categorical variables such as Sex be encoded before applying machine learning algorithms?

a: Because machine learning algorithms require numerical input Explanation

Explanation

b: Because encoding reduces dataset size Explanation

Explanation

c: Because encoding removes missing values Explanation

Explanation

d: Because encoding automatically improves model accuracy Explanation

Explanation

Which potential issue may arise when using One-Hot Encoding for categorical variables with many unique categories?

a: Feature dimensionality may increase significantly Explanation

Explanation

b: Numerical values may become negative Explanation

Explanation

c: Missing values automatically disappear Explanation

Explanation

d: The dataset size decreases Explanation

Explanation

Why is normalization applied to numerical features during preprocessing?

a: To convert categorical variables into numerical form Explanation

Explanation

b: To bring numerical features to a comparable scale Explanation

Explanation

c: To remove missing values Explanation

Explanation

d: To reduce the number of features Explanation

Explanation

Which type of machine learning algorithm is most sensitive to feature scaling?

a: Distance-based algorithms such as K-Nearest Neighbors Explanation

Explanation

b: Decision tree algorithms Explanation

Explanation

c: Rule-based systems Explanation

Explanation

d: Database query systems Explanation

Explanation

Which visualization technique is most suitable for identifying outliers in numerical data?

a: Histogram Explanation

Explanation

b: Box plot Explanation

Explanation

c: Pie chart Explanation

Explanation

d: Bar chart Explanation

Explanation

Why is data visualization performed after preprocessing in this experiment?

a: To verify whether preprocessing steps were applied correctly Explanation

Explanation

b: To automatically train a machine learning model Explanation

Explanation

c: To increase dataset size Explanation

Explanation

d: To remove duplicate values Explanation

Explanation

After completing preprocessing and feature engineering, what is the next logical step in a machine learning workflow?

a: Model training and evaluation Explanation

Explanation

b: Data collection Explanation

Explanation

c: Dataset labeling Explanation

Explanation

d: Feature deletion Explanation

Explanation