Generation and Termination of Binary Decision Trees for Nonparametric Multiclass Classification

摘要

A two-step procedure for nonparametric multiclass classifier design is described. A multiclass recursive partioning algorithm is given which generated a single binary decision tree for classifying all classes. The algorithm minimizes the Bayes risk at each node. A tree termination algorithm is given which optimally terminates binary decision trees. The algorithm yields the unique tree with fewest nodes which minimizes the Bayes risk. Tree generation and termination are based on the training and test samples, respectively.