PRODUCTION SCENARIO
A retailer has historical orders labeled as returned or kept. It wants to predict whether a new order will be returned.
Which approach will meet these requirements?
Answering here is anonymous. Nothing is saved unless you sign in.
Show answer and explanation
Answer: Supervised binary classification
Every historical order carries a known label with exactly two values, so a supervised algorithm learns the mapping and predicts one of two categories. Multiclass classification is the same family but assumes three or more outcomes, and there are only two here. Regression predicts a continuous number, not a category. Clustering discovers groups without labels, which throws away the returned-or-kept labels the retailer already has.
Official Source: https://docs.aws.amazon.com/sagemaker/latest/dg/algos.html ↗