Tree Classifiers

Abstract

Classification trees partition R ^d into regions, often hyperrectangles parallel to the axes. Among these, the most important are the binary classification trees, since they have just two children per node and are thus easiest to manipulate and update. We recall the simple terminology of books on data structures. The top of a binary tree is called the root. Each node has either no child (in that case it is called a terminal node or leaf), a left child, a right child, or a left child and a right child. Each node is the root of a tree itself. The trees rooted at the children of a node are called the left and right subtrees of that node. The depth of a node is the length of the path from the node to the root. The height of a tree is the maximal depth of any node.