Google Cloud Professional Machine Learning Engineer — Question 146

You are working on a binary classification ML algorithm that detects whether an image of a classified scanned document contains a company’s logo. In the dataset, 96% of examples don’t have the logo, so the dataset is very skewed. Which metric would give you the most confidence in your model?

Answer options

• A. Precision
• B. Recall
• C. RMSE
• D. F1 score

Correct answer: D

Explanation

The F1 score is the most appropriate metric in this scenario because it balances precision and recall, which is crucial in a skewed dataset. Precision alone could be misleading when there are many true negatives, and recall alone may not reflect the model's accuracy in identifying logos. RMSE is not relevant for classification tasks, making it an unsuitable choice.