1. . Model evaluation in machine learning is:
(A) Encrypting models only
(B) The process of assessing how well a trained model performs on unseen data
(C) Compressing datasets only
(D) Backup only
2. . Validation of a model is done to:
(A) Tune hyperparameters and check performance before final testing
(B) Encrypt validation data
(C) Compress features
(D) Backup only
3. . Training set is used to:
(A) Fit the model by learning patterns from data
(B) Encrypt data
(C) Compress data
(D) Backup only
4. . Test set is used to:
(A) Compress test set
(B) Encrypt test data
(C) Evaluate the model's performance on unseen data
(D) Backup only
5. . Cross-validation is:
(A) A technique to assess model performance by dividing data into multiple training and validation sets
(B) Encrypting cross data
(C) Compressing validation sets
(D) Backup only
6. . K-Fold cross-validation divides data into:
(A) Encrypts K folds
(B) Two parts only
(C) K subsets where each subset is used as a validation set once
(D) Backup only
7. . Leave-One-Out Cross-Validation (LOOCV) is:
(A) Encrypting data points
(B) Using one data point as validation and the rest for training, repeated for all points
(C) Compressing data points
(D) Backup only
8. . Accuracy is:
(A) Backup only
(B) Encrypting accuracy
(C) Compressing accuracy
(D) The ratio of correctly predicted instances to total instances
9. . Precision measures:
(A) Backup only
(B) Encrypting precision
(C) Compressing precision
(D) The proportion of true positives among all predicted positives
10. . Recall (Sensitivity) measures:
(A) Backup only
(B) Encrypting recall
(C) Compressing recall
(D) The proportion of true positives among all actual positives
11. . F1-Score is:
(A) Encrypting F1
(B) The harmonic mean of precision and recall
(C) Compressing F1
(D) Backup only
12. . Confusion matrix shows:
(A) Backup only
(B) Encrypting matrix
(C) Compressing matrix
(D) True positives, false positives, true negatives, and false negatives
13. . ROC Curve is used to:
(A) Evaluate the trade-off between true positive rate and false positive rate
(B) Encrypt ROC
(C) Compress ROC
(D) Backup only
14. . AUC (Area Under Curve) represents:
(A) The model's ability to discriminate between classes
(B) Encrypting AUC
(C) Compressing AUC
(D) Backup only
15. . Mean Squared Error (MSE) is used for:
(A) Compressing errors
(B) Encrypting errors
(C) Evaluating regression models by measuring average squared difference between actual and predicted values
(D) Backup only
16. . R-squared (Coefficient of Determination) measures:
(A) Proportion of variance in the dependent variable explained by the model
(B) Encrypting R-squared
(C) Compressing R-squared
(D) Backup only
17. . Overfitting occurs when:
(A) Compressing model
(B) Encrypting overfitting
(C) A model performs well on training data but poorly on unseen data
(D) Backup only
18. . Underfitting occurs when:
(A) Encrypting underfitting
(B) A model is too simple to capture underlying patterns
(C) Compressing model
(D) Backup only
19. . Early stopping in model validation helps to:
(A) Backup only
(B) Encrypt training
(C) Compress iterations
(D) Prevent overfitting by stopping training when performance on validation set stops improving
20. . The main purpose of model evaluation and validation is to:
(A) Ensure the model generalizes well, is accurate, and reliable before deployment
(B) Encrypt models
(C) Compress data
(D) Backup only