2
$\begingroup$

In machine learning, regression algorithms attempt to estimate the mapping function (f) from the input variables (x) to numerical or continuous output variables (y).

I have one usecase where I predict shift_id. Shit_Id is ID values given to different city location.

As per my understanding this is regression problem because it predict numerical value. Is this right?

Also precision, recall f1 measure can be calculated for regression problem?

$\endgroup$
1
$\begingroup$

IDs are categorical, not numeric. You should be treating this as a multi-class classification problem. Your IDs are locations, a location is a class. The ID is just a identifier for the class.

Since you have a classification problem you should be using precision, recall and f1. However, if it was regression you would have been using mean squared error, mean absolute error and possibly something else.

| improve this answer | |
$\endgroup$
  • $\begingroup$ Thanks for such a clear answer $\endgroup$ – Jhon Patric Apr 2 '19 at 7:42
  • $\begingroup$ I dont have enough points to upvote your answer $\endgroup$ – Jhon Patric Apr 2 '19 at 7:42
  • $\begingroup$ Don't worry about that. Glad I could help. Good luck! :) $\endgroup$ – Simon Larsson Apr 2 '19 at 7:44

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.