In a Spark MLlib project, you are tasked with comparing the performance of linear regression and decision tree models on a large dataset. Which of the following evaluation metrics would be most appropriate for comparing the performance of these models, and why?

Simulated

Accuracy, as it measures the proportion of correct predictions made by the model.

13.4%

Mean squared error (MSE), as it provides a measure of the average squared difference between the predicted and actual values.

56.7%

Area under the receiver operating characteristic (ROC) curve, as it measures the model's ability to distinguish between different classes.

12.7%

R-squared, as it measures the proportion of the variance in the target variable that is explained by the model.

17.2%

Databricks Certified Machine Learning - Associate