Monitoring and model drift
Monitoring is an important piece of a full AI solution. It enables the team to know when the AI algorithm may be at risk of decreased performance, or model drift.
The example below shows customer satisfaction scores pre-pandemic (teal blue) and post-pandemic (dark orange). The average score pre-pandemic is 2 while the average post-pandemic is 3.
Based on the descriptions of model drift in the video, how would you describe what is happening to customer satisfaction?
This exercise is part of the course
Implementing AI Solutions in Business
Hands-on interactive exercise
Turn theory into action with one of our interactive exercises
