1
$\begingroup$

I'm a newcomer to machine learning and currently diving into supervised learning methods. I've already grasped the theoretical basics of classification tasks and have just started exploring regression. While reading Sebastian Raschka's book, this question popped into my head. After Google failed to provide any answers, I decided to ask it here.

$\endgroup$

1 Answer 1

2
$\begingroup$

From a somewhat technical perspective, this is roughly correct: you can approximately solve a regression problem by solving the classification problem that you mentioned, but ...

... there is a conceptual difference that is more subtle: a class is of categorical type which implies that classes do not necessarily have a specific ordering, however, ordering is exactly what you need for a regression problem -- and which you probably implicitly assumed.

$\endgroup$
1
  • $\begingroup$ Agree, however, you can do classification where the labels have some order and you can take this into account in your loss, see for instance this question $\endgroup$ Commented Apr 6 at 12:14

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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