You have developed a spam detection system. It is more important to ensure that legitimate emails do not get classified as spam (minimizing false positives) than it is to catch every single spam email. Which evaluation metric should be your primary focus?

• Accuracy

• Precision

• Recall

• F1-score

You are working on a medical diagnosis model for a rare but serious disease. It is crucial to identify as many actual cases as possible, even if it means some healthy individuals might be flagged for further testing. Which metric is most important in this scenario?

• Accuracy

• Precision

• Recall

• F1-score

You have trained two different binary classification models for the same problem, and their ROC curves are plotted. One curve is consistently closer to the top-left corner of the plot than the other. Which model likely has better overall discriminatory power across various thresholds?

• The model with the ROC curve closer to the bottom-right corner.

• The model with the ROC curve closer to the top-left corner.

• The model whose ROC curve is closer to the diagonal line.

• It's impossible to tell from the ROC curve alone.

You are comparing the performance of several machine learning models on an imbalanced dataset. One model has a very high accuracy (e.g., 98%), but upon examining the confusion matrix, you find that it rarely correctly identifies the minority class. Which metric would provide a more informative assessment of the model's performance on the minority class?

• Accuracy

• Recall

• Number of true negatives

• Total number of samples

You are building a linear regression model to predict house prices based on square footage. You want to understand how much of the variance in house prices can be explained by the square footage. Which metric would directly answer this question?

• Mean Absolute Error (MAE)

• Root Mean Squared Error (RMSE)

• R-squared

• Pearson correlation

You have added several new features to your linear regression model, and the R-squared value has increased. However, the adjusted R-squared has decreased. What does this suggest?

• The new features have significantly improved the model's fit.

• The new features do not add significant explanatory power and may be overfitting the model.

• The model's predictive accuracy has decreased.

• There is a strong negative correlation between the new features and the target variable.

You are evaluating the performance of a regression model that predicts the time it takes for a customer to receive an order. You notice that the model occasionally makes very large errors. Which error metric would be most sensitive to these large errors?

• Mean Absolute Error (MAE)

• Root Mean Squared Error (RMSE)

• Both MAE and RMSE equally

• Neither MAE nor RMSE

You are comparing two regression models predicting the same continuous variable. Model A has a lower MAE than Model B, but Model B has a lower RMSE than Model A. What can you infer about the types of errors each model is making?

• Model A has larger errors, while Model B has more consistent smaller errors.

• Model B has larger errors, while Model A has more consistent smaller errors.

• Both models have similar error distributions.

• It is impossible to infer the error types from MAE and RMSE alone.

You are working on a multi-class classification problem with three equally important classes. You want a single metric that summarizes the overall performance of your model across all classes, giving equal weight to each class's performance in terms of precision and recall. Which of the following would be a suitable metric?

• Accuracy

• Precision for a single class

• Recall for a single class

• Macro-averaged F1-score

According to the sources, machine learning is primarily concerned with:

• Designing and building physical robots.

• The theoretical study of computation.

• The automated recognition and/or detection of meaningful and insightful patterns in data.

• Developing new programming languages.