Databricks Machine Learning Associate Question 250
Select 3During the evaluation of a binary classification model in Databricks, you notice that the dataset is highly imbalanced, with a significantly smaller number of positive instances compared to negative instances. Which of the following statements about Recall and F1 Score is/are true in this context?
- A
Recall is a better metric than Precision for this scenario because it focuses on the model's ability to identify positive instances.
- B
F1 Score is a good choice for evaluation when there is a class imbalance, as it balances the trade-off between Precision and Recall.
- C
Recall is independent of class imbalance and is not affected by the distribution of positive and negative instances.
- D
F1 Score will always be higher than Recall when class imbalance is present.
- E
Optimizing for Recall might lead to a higher number of false positives.
Show answer and explanation
Correct answers: A, B, E
Explanation
In imbalanced datasets, Recall is particularly useful for measuring how well the model identifies positive instances. However, focusing solely on Recall might lead to more false positives. F1 Score is often used in such scenarios as it balances Recall and Precision, making it a suitable choice for imbalanced classification problems. The other options reflect common misconceptions or incorrect interpretations of these metrics.
- A. Correct.
Recall is a key metric in imbalanced datasets since it measures the ability of the model to correctly identify positive instances, which are often the minority class in imbalanced datasets.
- B. Correct.
F1 Score combines both Precision and Recall, making it a robust choice for imbalanced datasets where both metrics are important.
- C. Incorrect.
Recall is affected by class imbalance because it depends on the ability to identify true positives, which may be harder to achieve in imbalanced datasets.
- D. Incorrect.
F1 Score does not always exceed Recall. It depends on the relative values of Precision and Recall, as it is the harmonic mean of the two.
- E. Correct.
Optimizing for Recall can increase the model's sensitivity to positive instances, potentially leading to more false positives.