Many technologies can be used to help or automate decision making. Business rules and decision management is a prominent category, but there are others, including machine learning and optimization. While seemingly unrelated, machine learning, decision optimization, and business rules can be used in combination to provide unprecedented accuracy in decision making. We will discuss several compelling examples of such combination used in various industries in the first part of this talk.
Then, we will review machine learning and decision optimization in more details. Both are based on mathematical optimization algorithms: a business problem is cast into an optimization problem. However, the problem to be optimized is very different. For machine learning, the problem is to minimize the difference between predictions and observed target values. For decision optimization, the problem is to minimize some operations cost. We will also show how data scientists approach machine learning and decision optimization in practice via the use of declarative modeling tools.