A two‐phase case‐based distance approach for multiple‐group classification problems

Classification is a procedure to separate data or alternatives into two or more classes. In practice, the need to classify alternatives involving multiple criteria into distinct classes is considerable. Therefore, determining how to assist decision makers in classifying alternatives into multiple classes is an important issue in the field of multiple‐criteria decision aids. This study proposes a two‐phase case‐based distance approach used to assist decision makers to classify alternatives into multiple groups. By incorporating the advantages of the case‐based distance method, the proposed two‐phase approach can classify alternatives by evaluating a set of cases selected by decision makers, reduce the number of misclassifications, improve multiple solution problems, and lessen the impact of outliers. An interactive classification procedure is also proposed to provide flexibility in such a way that decision makers can check and adjust classification results iteratively.