Extracting Purchase Patterns in Convenience Store E-Commerce Market Using Customer Cube Analysis

摘要

This paper proposes a method to capture the purchase patterns of customers by analyzing purchase history data in time series. This type of data is analyzed conventionally with two axes of Merchandise and Customer. In this paper, we use an analytical method called "Customer Cube", which analyze the data in three-dimensions of Merchandise, Customer, and Time axes. As a result of analyses using the Customer Cube, we have extracted three types of purchase patterns from the purchase history data on a CVS EC site. By capturing the transitions of customer preferences based on the extracted purchase patterns, CVSs should be able to increase the efficiency of their marketing, merchandising and merchandise management activities, which eventually contributes to increase the sales on their EC sites.