Here's a quite common problem and I read a couple of questions/answers on it, however I still having my doubts about what are the best practices for grouping data by Customer ID for churn prediction.

I have a dataset with CustomerID with weekly occurrences (so, there any Customer ID can have anywhere between 1 to let's say 26 occurrences). And every record comes with approx. 20 features, some of which are categorical.

The question is, what is the better approach to grouping data by ID: 1) Takes means/st.devs/whatever for numerical features and encode categorical ones, potentially losing some useful information. Or 2) Create some kind of temporally organized features (e.g. by week), which would be not that straightforward thing to do due to sheer difference in Customers' rate of occurrences. Also this would significantly increase the feature space. Or 3) Other suggestions?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.