For example, let's say that the age (say $x$) of a person for $12< x< 25$ can be used to predict computer usage to a high degree of certainty. In a decision tree, this could be represented by a split of $x<25$ followed by a split of $x>12$.

Does xgboost, or other decision tree learning algorithms in general, handle multiple splits within the same branch?


Yes, it can handle multiple splits withing the same branch.
A decision tree model can use the same feature as many times as optimally needed.
You can see an example in the sklearn doc.
The petal-length feature is used multiple times in the same branch.

| improve this answer | |

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.