Working with ML insights

QuickSight has a special type of insight where the results are driven from ML-based computations. QuickSight supports two types of ML insights:

• Forecasting
• Anomaly detection

In the next section, we will learn how to configure each of these types of insights.

Working with forecasting insights

We learned how to add forecasting in a line graph visual. QuickSight also allows you to add forecasting as a narrative to display forecasted values. For example, how many miles is expected to be traveled between Manhattan and Queens on a specific date?

To better understand how to configure this type of visual, we will use our example New York Taxi analysis to answer this question:

1. First, let's create a new insight and select Forecast ML-Powered Insight from the drop-down list.
2. For this example, let's assume that we want the user to choose the pick-up borough and the destination borough (drop-off borough). To achieve...