T4Tutorials .PK

NSCT – Model Evaluation & Validation MCQs

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




Exit mobile version