DATA QUALITY IN THE CONTEXT OF CUSTOMER SEGMENTATION

摘要

There's no such thing as data quality in general: it all depends on the particular business application. With this in mind, we begin by introducing customer segmentation as a basic data mining application in analytical CRM. We describe the structure of the data mining environment and outline data selection as an essential, but difficult sub-problem. From there, we go on to develop three perspectives for approaching the issue: marketing, data mining, and data quality. Then we introduce a process model for the systematic identification of input variables for customer segmentation. Additionally, we offer practical suggestions and set out difficulties and requirements for further development.