2
$\begingroup$

I have been reading a book about Decision Trees and it caught my attention the following part:

In case of numeric attributes, decision trees can be geometrically interpreted as a collection of hyperplanes, each orthogonal to one of the axes

I tried to look for information about this in Internet, but I did not find anything. To what the author refers in this point?

$\endgroup$
2
$\begingroup$

Each vertex in a decision tree is associated with a question of the form "$x_i < c$?" (or "$x_i \leq c$?"), where $x_i$ is one of the input and $c$ is a constant. The decision whether to go left or right then depends on which side of the hyperplane $x_i = c$ the input is located.

The possible hyperplanes which come up this way are not arbitrary – they are orthogonal to one of the axes; the hyperplane $x_i = c$ is orthogonal to the $x_i$ axis. An example of a different hyperplane, which cannot occur, is $x_1 - x_2 = 0$.

$\endgroup$
  • $\begingroup$ thanks @Yuval Filmus, could you be so kind to recommend me some bibliography about that in a comment? Thanks $\endgroup$ – Layla Nov 2 '18 at 16:53
  • $\begingroup$ I can't think of anything in particular, unfortunately. $\endgroup$ – Yuval Filmus Nov 2 '18 at 16:55

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.