I have a dataset which has around 10K records.

My objective is to predict whether the customer will churn or not. Binary classification problem with each class representing around 55:45 proportion and 20 features.

I understand when it's just about prediction, I can apply some binary classification algorithm and find out whether the customer churns or not

But how do I incorporate the objective of finding whether the customer will churn in 30 days or not?

Another example is find whether patient will be dead within 30 days from the date of discharge. I have his date of discharge along with other features like Blood pressure, Cholesterol etc.

Rather than just predicting whether he will be dead or not anytime in future, I would like to restrict it to 30 days from date of discharge.

Hope I gave the details to help you understand the question better.

  • $\begingroup$ Can you tell us some more about the structure of your dataset? You should look into survival analysis, or methods for recurrent events, see for instance stats.stackexchange.com/search?q=recurrent+events+answers%3A1 $\endgroup$ – kjetil b halvorsen Feb 17 '20 at 7:12
  • $\begingroup$ My dataset is structured dataset with 20 features (mix of continuous & discrete variables) and two possible outcomes. Churn or not churn $\endgroup$ – The Great Feb 17 '20 at 7:31
  • $\begingroup$ But do you have many observations for each customer? How many times do they typically recur in a 30-day interval? If they today have a much longet inteval since last seen, than their typical mean recurrence time, maybe they are about to churn ... $\endgroup$ – kjetil b halvorsen Feb 17 '20 at 7:35
  • $\begingroup$ Each customer has atleast more than 10 records. Is there any tutorial that you can share for this?? would really be helpful $\endgroup$ – The Great Feb 17 '20 at 7:53
  • $\begingroup$ I doubt there is a simple tutorial for your specific case. Can you share (a link to) the data (or some mockup data). Else look at the links for recurrent events. And, please add all new information as an edit to the Q. Not all people read commets. $\endgroup$ – kjetil b halvorsen Feb 17 '20 at 7:59

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.