PRODUCTION SCENARIO
A meal-delivery company wants to predict how many minutes each order will take to arrive. The training data includes distance, time of day, and restaurant preparation history. The target column holds values such as 27 and 41.
Which model type should the company use?
Answering here is anonymous. Nothing is saved unless you sign in.
Show answer and explanation
Answer: Numeric prediction (regression)
The target is a continuous number of minutes, which is regression, called numeric prediction in Amazon SageMaker Canvas. The category prediction types would force the minute values into buckets and discard the quantity being asked for, and time series forecasting predicts future values of a series over time rather than an attribute of a single order.