온라인 소매상점에서의 효과적인 고객 분류 방법론: 의사결정나무 기법에의 적용

On-line retailers are introducing a customized services to obtain competitive advantage in an increasingly tighter electronic marketplace. Understanding customers’ preferences for products and administering appropriate marketing strategies are essential for a personalization. Marketing researchers and practitioners have focused on this issue and are tackling the problem of customer classification beyond the bounds of traditional market segmentation in the relatively new but rapidly disseminating environments afforded by advances in information technology. Application of existing decision tree algorithms for on-line retail customer classification is apt to construct a bushy tree due to noise source data like click stream. In this study, we propose an effective methodology for classifying on-line customers when the decision trees is used to a segmenting technique. Basic idea of the study is to determine the appropriate number of segments with a certain level of abstraction. We developed a stopping rule that considers the total amount of information gained while generating a rule tree. In addition to forwarding from root to intermediate nodes with a certain level of abstraction, the decision tree is investigated by the backtracking pruning method with misclassification loss information.