Data and Analytics Machine Learning Fundamentals 3

Question 1

Which metric is most appropriate when evaluating a classification model on a highly imbalanced dataset?

Accepted Answer

F1-score

Answer

Accuracy

Answer

Mean squared error

Answer

R-squared

Question 2

What does an ROC curve illustrate in model evaluation?

Accepted Answer

The tradeoff between true positive rate and false positive rate at various thresholds

Answer

The relationship between training loss and validation loss

Answer

The distribution of residuals in a regression model

Answer

The importance of each feature in a classification model

Question 3

In k-fold cross-validation, what happens to the data?

Accepted Answer

The data is divided into k equal parts; each fold serves as the test set once while the rest train the model

Answer

The data is split once into training and test sets

Answer

The model is trained k times on the full dataset

Answer

The data is bootstrapped k times to create new training sets

Question 4

What is a confusion matrix used to measure?

Accepted Answer

The counts of true positives, true negatives, false positives, and false negatives

Answer

The correlation between input features

Answer

The variance explained by each principal component

Answer

The convergence rate of the training algorithm

Question 5

Which of the following best describes hyperparameter tuning?

Accepted Answer

Searching for the best configuration settings that control the learning process

Answer

Adjusting model weights during backpropagation

Answer

Removing irrelevant features from the dataset

Answer

Increasing the size of the training data

Question 6

What is the primary advantage of using ensemble methods like bagging?

Accepted Answer

They combine multiple models to reduce variance and improve predictive performance

Answer

They reduce the number of hyperparameters to tune

Answer

They eliminate the need for feature engineering

Answer

They train faster than single models on large datasets

Question 7

Which approach is used to handle missing values in a dataset by replacing them with the mean, median, or mode?

Accepted Answer

Imputation

Answer

Feature scaling

Answer

Dimensionality reduction

Answer

One-hot encoding