FUZZY TARGET SELECTION IN DIRECT MARKETING

S. Ramathilagam (Taiwan) and S. R. Kannan (India)

Received February 11, 2008

Abstract

This work concerns some clustering algorithms for target selection in direct marketing problem. The problem of clustering in data is concerned with finding groups or structures within a finite number of data. Target selection is the problem of finding groups of customers for a particular product in direct marketing. In the problem of direct marketing, manufacturing companies try to have a contact or maintain a direct relationship with customers in order to target them individually for a particular product offers or maximizing the profit. This work likes to present some fuzzy and k-means clustering methods to identify profiles of potential customers for direct marketing. This work reviews the target problems and different mathematical methods in finding potential customers for direct marketing. This work is planned to implement computer assisted cross-validation of multi-regression for data of target selection for direct marketing. Also then this work is planned to introduce fuzzy clustering methods and k-means for dividing the customer database into groups with similar properties called customer segments to maximize the profit or fund rising in direct marketing. Finally, the results from fuzzy clustering methods and k-means will be compared.

Keywords and phrases: computer assisted cross-validation, multi-regression, direct marketing, fuzzy clustering methods.