Are there any classification schemes like a support vector machine that use axis-aligned boxes instead of hyperplanes? I have a dataset consisting of about a billion points in 9 dimensions, which are divided into two classes. I would like a classification rule that gives me a collection of disjoint boxes, such that the rule for classifying query point $x$ is to say "if $x$ is in any of the boxes, it's class 1, and otherwise, it's class 2". Is this a thing?


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.