Fuzzy clustering in data mining for telco database marketing campaigns

Fuzzy methods have been applied to data mining and to databases of customer information for marketing. This paper explores fuzzy clustering approaches to telecommunications database marketing. Fuzzy clustering methods can be used to mine telco customer and prospect databases to gain residential and business customer market share. Four key fuzzy enhancements to traditional database marketing are developed in this paper. First, customers often have significant membership values in more than one distinct fuzzy cluster and can be considered in a natural manner for hybrid or multiple contacts in a given marketing campaign. Second, fuzzy clustering outcomes are shown to be dependent on the particular offer or marketing message. Third, there are differences in clustering outcomes over time, as various offers and treatments are successively presented to consumers, and as products and tastes change. This evolution of fuzzy clusters can be used to help understand customer loyalty and to extract a more optimal lifetime economic relationship value. Fourth, in the longer run, formal procedures can be suggested involving intuitive fuzzy-based clustering metrics for continuous process improvement, to support increasingly flexible and opportunistic campaign management.