1. . Supervised learning is:
(A) Training models with no data
(B) A type of machine learning where the model is trained on labeled data
(C) Compressing unlabeled data
(D) Backup only
2. . Labeled data in supervised learning contains:
(A) Only input features
(B) Input features and their corresponding output or target values
(C) Only outputs
(D) Backup only
3. . The main goal of supervised learning is to:
(A) Compress data
(B) Encrypt data
(C) Learn a mapping from inputs to outputs to make predictions on new data
(D) Backup only
4. . Regression in supervised learning is used to:
(A) Backup only
(B) Predict categories
(C) Encrypt numbers
(D) Predict continuous numeric values
5. . Classification in supervised learning is used to:
(A) Encrypt categories
(B) Predict continuous values
(C) Predict categorical or discrete outcomes
(D) Backup only
6. . Common supervised learning algorithms include:
(A) Principal component analysis only
(B) K-means clustering only
(C) Linear regression, logistic regression, decision trees, random forests, and support vector machines
(D) Backup only
7. . Mean Squared Error (MSE) is commonly used to:
(A) Backup only
(B) Encrypt errors
(C) Compress errors
(D) Evaluate regression models
8. . Accuracy, precision, recall, and F1-score are used to:
(A) Compress metrics
(B) Encrypt metrics
(C) Evaluate classification models
(D) Backup only
9. . Training set in supervised learning is:
(A) Encrypting data
(B) The subset of data used to train the model
(C) Compressing data
(D) Backup only
10. . Test set in supervised learning is:
(A) Compressing test data
(B) Encrypting test data
(C) The subset of data used to evaluate the model's performance on unseen data
(D) Backup only
11. . Overfitting occurs when:
(A) Compressing models
(B) Encrypting models
(C) The model performs well on training data but poorly on new data
(D) Backup only
12. . Underfitting occurs when:
(A) Compressing models
(B) Encrypting models
(C) The model is too simple to capture patterns in the data
(D) Backup only
13. . Cross-validation is used to:
(A) Encrypt data splits
(B) Assess model performance more reliably and prevent overfitting
(C) Compress validation data
(D) Backup only
14. . Feature scaling in supervised learning helps:
(A) Backup only
(B) Encrypt features
(C) Compress features
(D) Improve convergence and performance of algorithms sensitive to feature magnitude
15. . Decision trees in supervised learning:
(A) Split data based on feature values to make predictions
(B) Encrypt trees
(C) Compress trees
(D) Backup only
16. . Support Vector Machines (SVM) aim to:
(A) Encrypt hyperplanes
(B) Find the optimal hyperplane that separates classes in the feature space
(C) Compress feature spaces
(D) Backup only
17. . K-Nearest Neighbors (KNN) predicts output by:
(A) Considering the majority label or average value of nearest neighbors
(B) Encrypting neighbors
(C) Compressing neighbors
(D) Backup only
18. . Regularization in supervised learning is used to:
(A) Backup only
(B) Encrypt coefficients
(C) Compress models
(D) Reduce overfitting by penalizing large coefficients
19. . Label encoding is used to:
(A) Encrypt labels
(B) Convert categorical variables into numeric labels for modeling
(C) Compress labels
(D) Backup only
20. . The main purpose of supervised learning is to:
(A) Predict outputs for new inputs using a model trained on labeled data
(B) Encrypt all data
(C) Compress all features
(D) Backup only