3. (a) Briefly explain the notion of a profit curve.

(b) Compare and contrast profit curves with ROC curves i.e., describe their similarities and differences).

(c) Why might we choose to use a profit curve instead of an ROC curve? Why might we choose use an ROC curve instead of a profit curve?

(d) What can we conclude about a classifier if the area under its ROC curve is equal to 0.5?

(e) Suppose that you are a data scientist who works for the medical center in Yugoslavia where the breast cancer dataset used in Table I was collected. In your upcoming presentation to your bosses, you aim to convince them that the SVM model performs well and can save the medical center money. Select one or more performance metrics, graphs or curves to show at the meeting. Explain why you chose those metric(s) or chart(s) and what information you aim to convey with each of them.