MCQs

Disclaimer:

Educational Use Only:

The website links provided are to be used strictly for educational and knowledge-gaining purposes only.

No Misuse:

Do not misuse the links or the information contained therein for any commercial, illegal, or unethical activities.

Respect for Websites and Businesses:

Respect the intellectual property and business interests of the websites and their owners. Do not attempt to alter, copy, or redistribute content without proper authorization.

Acknowledge Value of Knowledge:

Recognize and value the knowledge and resources shared by these websites. Use the information responsibly and ethically.

No Endorsement:

The provision of these links does not imply endorsement of the linked sites, their content, or their products and services. Users are responsible for evaluating the suitability and accuracy of the information.

User Responsibility:

By accessing these links, you agree to use the information at your own risk and to comply with all applicable laws and website terms of use.

Overall MCQs for Reference

Which metric is most commonly used to measure classification model accuracy?
a. Mean Squared Error
b. Recall
c. Accuracy
d. R-squared
Answer: c. Accuracyappliedaicourse+1
What is Precision in model evaluation?
a. Proportion of all actual positives
b. Proportion of predicted positives that are actual positives
c. Ratio of False Positives to True Positives
d. Measure of variance
Answer: b. Proportion of predicted positives that are actual positivesgeeksforgeeks+1
What does Recall tell us in a classification problem?
a. Proportion of predicted negatives
b. Proportion of actual positives correctly identified
c. Proportion of incorrect predictions
d. Average of Precision and F1-score
Answer: b. Proportion of actual positives correctly identifiedappliedaicourse+1
The F1-score is the:
a. Arithmetic Mean of Precision and Recall
b. Product of Precision and Recall
c. Harmonic Mean of Precision and Recall
d. Maximum of Precision and Recall
Answer: c. Harmonic Mean of Precision and Recallgeeksforgeeks+1
Which metric is ideal for evaluating regression models with outlier sensitivity?
a. MAE
b. Accuracy
c. MSE
d. AUC-ROC
Answer: c. MSEneptune+2
What does AUC-ROC stand for?
a. Average Using Counts – Receiver Operating Curve
b. Area Under the Curve – Receiver Operating Characteristic
c. Accuracy Under Curve – Regression Output Calculation
d. Average Upper Calculation – Recall Only
Answer: b. Area Under the Curve – Receiver Operating Characteristicgeeksforgeeks+1
Which value in a confusion matrix represents correct positive predictions?
a. False Positive
b. False Negative
c. True Positive
d. True Negative
Answer: c. True Positivegeeksforgeeks
What is the purpose of cross-validation in machine learning?
a. Make model more complex
b. Test model stability and reduce overfitting
c. Select best features
d. Improve training speed
Answer: b. Test model stability and reduce overfittingstudocu+1
For highly imbalanced datasets, which metric is most informative?
a. Accuracy
b. Recall
c. Precision
d. F1-score
Answer: d. F1-scoreinterviewkickstart+1
Mean Absolute Error (MAE) is best described as:
a. Square root of errors
b. Average magnitude of errors, without direction
c. Percentage error
d. Fraction of correct predictions
Answer: b. Average magnitude of errors, without directionlabelstud+3
What does R-squared ( ) represent in regression evaluation?
a. Fraction of actual positives
b. Proportion of variance explained by the model
c. Proportion of predicted negatives
d. Rate of learning
Answer: b. Proportion of variance explained by the modelnature+3
Which technique helps detect model bias and variance problems visually?
a. ROC curve
b. Learning curve
c. Confusion matrix
d. Grid search
Answer: b. Learning curvestudocu
What does the holdout method do in model evaluation?
a. Uses all data for training
b. Splits data into training and testing sets
c. Only uses validation data
d. Ignores class imbalance
Answer: b. Splits data into training and testing setslearnovita+1
Which metric penalizes large errors more heavily in regression?
a. MAE
b. R-squared
c. MSE
d. F1 Score
Answer: c. MSEneptune+2
Which metric helps measure the fraction of incorrectly classified examples?
a. Loss
b. False Positive Rate
c. Error Rate
d. Recall
Answer: c. Error Ratedevinterview+1
What does k-fold cross-validation achieve?
a. Only increases model complexity
b. Randomly selects data for training
c. Provides better estimate of model generalization
d. Reduces dataset size
Answer: c. Provides better estimate of model generalizationlearnovita+1
Logarithmic Loss (Log Loss) is used for:
a. Regression models
b. Clustering models
c. Probabilistic classification models
d. Model explainability
Answer: c. Probabilistic classification modelsgeeksforgeeks
What is a confusion matrix used for?
a. Evaluating regression output
b. Summarizing prediction results in classification
c. Plotting data features
d. Increasing feature size
Answer: b. Summarizing prediction results in classificationgeeksforgeeks+1
Which is not a regression evaluation metric?
a. MAE
b. Recall
c. RMSE
d. MSE
Answer: b. Recallappliedaicourse+1
What does the term “overfitting” mean?
a. Model fits training data poorly
b. Model fits training and test data equally
c. Model fits training data well but fails on unseen data
d. Model performs better with more features
Answer: c. Model fits training data well but fails on unseen datastudocu+1
Which metric measures the proportion of true negatives out of all actual negatives?
a. Specificity
b. Recall
c. Precision
d. MAE
Answer: a. Specificitygeeksforgeeks+1
Why is normalization important before evaluating certain machine learning models?
a. It increases overfitting
b. Ensures features are on same scale for fair evaluation
c. Randomizes data
d. Reduces number of observations
Answer: b. Ensures features are on same scale for fair evaluationlabelstud+1
What kind of data split is used in the holdout validation technique?
a. Training and testing sets
b. Only training set
c. Only validation set
d. Testing and deployment sets
Answer: a. Training and testing setslearnovita+1
In binary classification, if both classes are equally important, which metric is best?
a. F1-score
b. ROC-AUC
c. Accuracy
d. Specificity
Answer: c. Accuracyappliedaicourse+1
Which validation technique splits data into k equally-sized folds?
a. Bootstrapping
b. Holdout split
c. K-fold cross-validation
d. Leave-one-out cross-validation
Answer: c. K-fold cross-validationstudocu+1
What metric is useful for evaluating clustering models?
a. Mean Squared Error
b. Silhouette Score
c. R-squared
d. Accuracy
Answer: b. Silhouette Scorepmc.ncbi.nlm.nih
What does statistical hypothesis testing in model evaluation involve?
a. Always increasing model complexity
b. Comparing model performance to a baseline
c. Splitting data randomly
d. Ignoring random errors
Answer: b. Comparing model performance to a baselinepmc.ncbi.nlm.nih+1
Which statement describes bootstrapping in validation?
a. Uses permutations without replacement
b. Resamples dataset with replacement to estimate statistics
c. Leaves out one data point per split
d. Only done after cross-validation
Answer: b. Resamples dataset with replacement to estimate statisticswowdata+1
Why is F1-score useful in class-imbalanced classification?
a. Only measures recall
b. Sensitive to both precision and recall, which are important when classes are imbalanced
c. Measures only accuracy
d. Ignores false positives
Answer: b. Sensitive to both precision and recall, which are important when classes are imbalancedinterviewkickstart+1
What is the advantage of k-fold cross-validation compared to holdout?
a. More computationally expensive
b. Utilizes all samples for both training and validation
c. Uses less data for validation
d. Skips bias detection
Answer: b. Utilizes all samples for both training and validationlearnovita+1
In which situation is high recall preferred over high precision?
a. False negatives are more harmful
b. False positives are more harmful
c. Overfitting is present
d. Balanced classes
Answer: a. False negatives are more harmfulgeeksforgeeks+1
Which metric is not suitable for multiclass classification?
a. Accuracy
b. F1-score (per class)
c. Specificity (per class)
d. Mean Squared Error
Answer: d. Mean Squared Errornature+1
Mean Squared Error (MSE) is calculated by:
a.
b.
c.
d.
Answer: b. nature+1
Which metric is best when prediction errors of all magnitudes are equally important?
a. RMSE
b. MAE
c. R-squared
d. Confusion matrix
Answer: b. MAElabelstud+3
AUC-ROC plots which values?
a. True Positive Rate vs. False Positive Rate
b. Accuracy vs. Loss
c. Precision vs. Recall
d. MAE vs. MSE
Answer: a. True Positive Rate vs. False Positive Rategeeksforgeeks+1
Which method splits data so every observation is used once for validation?
a. Random holdout
b. k-fold cross-validation
c. Leave-one-out cross-validation
d. Bootstrap aggregating
Answer: c. Leave-one-out cross-validationlearnovita
What type of error does RMSE penalize more than MAE?
a. Small errors
b. Random errors
c. Large errors
d. Balanced errors
Answer: c. Large errorsneptune+2
What does the confusion matrix NOT show?
a. True Positives
b. False Negatives
c. Prediction probabilities
d. False Positives
Answer: c. Prediction probabilitiesgeeksforgeeks+1
Why might accuracy be misleading for severely imbalanced classes?
a. Ignores false negatives and positives
b. Is most informative for imbalanced data
c. Considers only outliers
d. Always maximized
Answer: a. Ignores false negatives and positivesinterviewkickstart+1
Which evaluation is ideal for time-series forecasting?
a. Stratified sampling
b. Temporal (time-based) split validation
c. Random under-sampling
d. k-means clustering
Answer: b. Temporal (time-based) split validationlearnovita
Which test dataset is used to fine-tune model hyperparameters?
a. Testing set
b. Training set
c. Validation set
d. Data augmentation set
Answer: c. Validation setdevinterview
What is model bias?
a. Model learns very random, noisy patterns
b. Model performs consistently poorly due to incorrect assumptions
c. Model generalizes well
d. Model achieves best possible fit
Answer: b. Model performs consistently poorly due to incorrect assumptionsstudocu+1
Which metric combines both Sensitivity (Recall) and Specificity?
a. Precision
b. Balanced Accuracy
c. Error Rate
d. MAE
Answer: b. Balanced Accuracylabelstud+1
Which validation strategy is ideal for small datasets?
a. Holdout
b. k-fold cross-validation
c. Leave-one-out cross-validation
d. Bootstrapping
Answer: c. Leave-one-out cross-validationwowdata+1
What is the main purpose of performance metrics?
a. To increase overfitting
b. To prevent model evaluation
c. To quantify model effectiveness and enable comparison
d. To limit data usage
Answer: c. To quantify model effectiveness and enable comparisondevinterview+2
Which metric is suitable for unsupervised clustering evaluation without ground truth?
a. Silhouette Index
b. Accuracy
c. Log Loss
d. F1-score
Answer: a. Silhouette Indexpmc.ncbi.nlm.nih
The Wilcoxon signed-rank test is used for:
a. Comparing two related samples’ medians
b. Calculating MSE
c. Balancing classes
d. Determining feature importance
Answer: a. Comparing two related samples’ medianslearnovita
What is model variance?
a. Model performs poorly because it learns noise in data
b. Model always predicts the same output
c. Model underfits data
d. Model stability across training runs
Answer: a. Model performs poorly because it learns noise in datastudocu+1
What does high precision but low recall imply?
a. Most positive predictions are correct, but model misses many actual positives
b. Model catches all positives but with many false alarms
c. Model achieves high F1-score always
d. Model is always overfitting
Answer: a. Most positive predictions are correct, but model misses many actual positivesappliedaicourse+1
What is the main goal of model evaluation in machine learning?
a. To find errors in data
b. To measure how well the model generalizes to unseen data
c. To increase model complexity
d. To reduce computation time
Answer: b. To measure how well the model generalizes to unseen dataappliedaicourse+2

Blog

Systemethic Technologies

Leave a Reply Cancel reply

Useful links

Other Terms & Policies

Contact Info