Exam AWS Certified Big Data - Specialty topic 1 question 57 discussion

my selection C

it's yearly subscription; and the time series data is only for 3 months. Is this not a mis-match ? We should at least be reviewing a years data to make prediction ?

The ask is "On which basis should this binary classification model be built". To build a model we need recent utilization data. I will go with C

A. User profiles. https://aws.amazon.com/blogs/machine-learning/predicting-customer-churn-with-amazon-machine-learning/ Mobile operators have historical records on which customers ultimately ended up churning and which continued using the service. We can use this historical information to construct an ML model of one mobile operator’s churn using a process called training. After training the model, we can pass the profile information of an arbitrary customer (the same profile information that we used to train the model) to the model, and have the model predict whether this customer is going to churn.

to predict churn, we need to have user pattern for the last period. profile info can be added also

a? thoughts?