On tree-based classifications with multi-response variables

Data mining has recently received much attention in a wide range of business and engineering field. Classification is one of the most important topics in data mining. Tree-based approach, for example decision tree, is a very useful technique for finding classification models. Most researches on decision trees have been conducted for the case of single response variable. However, situations where multi-response variables should be considered arise from many areas such as process monitoring, marketing science, and clinical analysis. This paper is concerned with tree-based classification methods when there are two or more response variables in the data set. We first give an overview of tree-based approaches and then provide three kinds of node splitting methods for the case in which multi-response variables are of concern. An illustrative example of tree-based classifications is also given with discussion.