Leveraging ML for control flow
We have only scratched the surface of what can be accomplished with control services. Using rules to implement orchestration and CEP is clean and very powerful, but it is not the end of the road. We can certainly implement control flow with raw bespoke logic as well, but a very interesting approach that is emerging is the use of ML to steer control flow. For example, a control service could raise alerts based on facial recognition or fraud and anomaly detection, or a control service could generate leads and personalized recommendations based on user activity.To leverage ML, we need to look at both sides of the equation: models and predictions. Let's look at these in turn.
Models
When it comes to ML, it is all about the data. You need data, data, and more data. The more data you have, the more accurate your models will be. Conversely, if you do not have enough data, then you will most likely be better off using rules instead of ML.Fortunately, we have...