Abstract

This paper presents a data mining study that aims to identify potential high-value visitors for a drugstore chain in Japan. Our purpose is to provide timely decision support to the marketing and service departments for managing customer relationship. The conceptualization of customer value is discussed and is differentiated from a more commonly used construct, customer loyalty. We briefly describe the data mining system that supports the study. Our result show two supervised learning methods are comparable in terms of predictive accuracy.

Share

COinS