0
$\begingroup$

I'm creating a churn model.

My first thought was that the bigger the training set, it would be better.

However, 2020 was a crazy year because the COVID 19. For example, a user who was sick and isolated, had more free time, and as a result, spent more time on the APP that I'm analyzing. Therefore, many of the relevant features, have different range of values compared to their values for users that used the app in 2022.

A simple and immediate solution is to shorten the training period. However, a lot of available data will not be used.

Is there a solution to help the model handle the old data without removing it?

$\endgroup$

0

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.