Non-actionable models
Identifying failed experiments and models that have not helped with driving the desired business outcomes is important: it helps ensure resources are allocated to the areas with the most business impact. Below are three results from a churn prevention test based on the ML model output. Which one performed the best and should be chosen for the implementation in the main production systems?
This exercise is part of the course
Machine Learning for Business
Hands-on interactive exercise
Turn theory into action with one of our interactive exercises
