Answer-first summary for fast verification
Answer: F-score
The F-score is recommended as it represents the harmonic mean of precision and recall, providing a single metric to evaluate the overall performance of classification models. Metrics like root mean squared error and mean squared error are more suited for regression models, not classification. Feature crosses are a method for creating synthetic features and do not serve as a performance metric. For more details, refer to [Google Cloud's AutoML Tables documentation](https://cloud.google.com/automl-tables/docs/evaluate).
Author: LeetQuiz Editorial Team
Ultimate access to all questions.
In the context of developing a machine learning model to classify potentially fraudulent transactions, where the goal is to rank models based on both precision and recall, which evaluation metric would you suggest?
A
Root mean squared error
B
F-score
C
Mean squared error
D
Feature crosses
No comments yet.