There are four key steps in the Automated Machine Learning (AutoML) process: data preparation, model building, creating ensemble models, and model recommendation. When examining or reporting on a model, there are many questions that can arise that are related to each of these steps. This activity is important because it will demonstrate your understanding of what questions might arise when looking at or reporting on the AutoML model's decision or recommendation. The goal of this activity is to anticipate and understand that questions should be considered or might be asked upon examining or reporting results of an Automated Machine Learning model. For each question, determine the correct step in the AutoML process where the question might arise. 1. How was the data collected for the analysis? 2. What are the reasons behind why the recommended model produced the most accurate decision? 3. What features should be used to build the model? 4. What problems might have existed before combining favorable elements from all models into a single model? 5. How was the data prepared for the analysis? 6. How did the model arrive at a particular conclusion? 7. Could a more accurate model be developed from combining different algorithms from more than one model? 8. What is the dependent variable of interest? 9. What variables had the greatest impact on the predicted outcome