1. . Ensemble learning is:
(A) Encrypting models
(B) A machine learning technique that combines multiple models to improve overall performance
(C) Compressing datasets
(D) Backup only
2. . The main goal of ensemble learning is to:
(A) Reduce errors, variance, and bias compared to individual models
(B) Encrypt predictions
(C) Compress predictions
(D) Backup only
3. . Bagging in ensemble learning stands for:
(A) Balanced Algorithm
(B) Binary Aggregation
(C) Bootstrap Aggregating
(D) Backup only
4. . Bagging improves model performance by:
(A) Compressing predictions
(B) Encrypting models
(C) Training multiple models on random subsets of data and averaging predictions
(D) Backup only
5. . Random Forest is:
(A) Encrypting trees
(B) An ensemble of decision trees using bagging and feature randomness
(C) Compressing trees
(D) Backup only
6. . Boosting is:
(A) Backup only
(B) Encrypting boosting
(C) Compressing boosting
(D) An ensemble method that trains models sequentially, giving more weight to previously misclassified instances
7. . AdaBoost stands for:
(A) Adaptive Boosting
(B) Automatic Boosting
(C) Algorithmic Boosting
(D) Backup only
8. . Gradient Boosting works by:
(A) Encrypting gradient
(B) Optimizing a loss function by sequentially adding models to correct errors
(C) Compressing gradients
(D) Backup only
9. . Stacking in ensemble learning is:
(A) Compressing predictions
(B) Encrypting stacks
(C) Combining predictions of multiple models using a meta-model
(D) Backup only
10. . Voting classifiers in ensemble learning:
(A) Encrypt votes
(B) Make predictions based on majority voting from multiple models
(C) Compress votes
(D) Backup only
11. . Ensemble methods are used to:
(A) Compress datasets
(B) Encrypt models
(C) Increase accuracy and robustness of machine learning models
(D) Backup only
12. . Key advantage of ensemble learning is:
(A) Encrypting predictions
(B) Reducing overfitting and improving generalization
(C) Compressing features
(D) Backup only
13. . Random Forest handles overfitting by:
(A) Using multiple decision trees and averaging their predictions
(B) Encrypting trees
(C) Compressing trees
(D) Backup only
14. . Difference between bagging and boosting is:
(A) Bagging encrypts data
(B) Bagging trains models in parallel, boosting trains sequentially
(C) Boosting compresses features
(D) Backup only
15. . Out-of-bag (OOB) error is used in:
(A) Encrypting error
(B) Random Forest to estimate prediction error without separate test data
(C) Compressing OOB data
(D) Backup only
16. . Ensemble methods are particularly useful when:
(A) Backup only
(B) Encrypting models
(C) Compressing datasets
(D) Individual models have high variance or bias
17. . Weighted voting in ensemble learning:
(A) Assigns different importance to each model's prediction
(B) Encrypts weights
(C) Compresses votes
(D) Backup only
18. . Bagging reduces:
(A) Data size
(B) Bias
(C) Variance of model predictions
(D) Backup only
19. . Boosting reduces:
(A) Backup only
(B) Variance only
(C) Data size
(D) Bias and improves model accuracy
20. . The main purpose of ensemble learning is to:
(A) Encrypt all models
(B) Combine multiple models to create a stronger, more accurate, and robust predictive model
(C) Compress all features
(D) Backup only