I'm working on a classification model for predicting customer behaviour. For each customer the dataset has multiple rows; each row representing information about his transaction during a quarter.

I would like to use decision trees but I don't really see how to aggravate the data (using mean or similar functions) so I have one row per customer.

  • $\begingroup$ You write that each row should represent the information of a given customer during a given quarter, so what exactly is your question? Is it about how to get your data into this format? If so, how should one answer that without knowing the 'is-state' of your data? $\endgroup$ – deemel Apr 9 at 14:33
  • $\begingroup$ @Rickyfox The dataset is in that format but I haven't found much useful literature on how to deal with that format. I would like to know if there are techniques that "flatten" the data. The point is that an instance is described by multiple rows (there's the customer id and transaction id which are the key for the dataframe) so the model shouldn't look at that rows separately. $\endgroup$ – Waddles Apr 10 at 8:15

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.