Watch this video on YouTube
A project sponsor wants to develop a machine learning model to classify potentially fraudulent transactions. They want to rank models based on a combination of precision and recall. What evaluation metric would you recommend?