Clustering and rule induction for Customer behavior analysis using Rough sets

Abstract

Customer behavior analysis is an important task in Customer newlineRelationship Management CRM because of the dynamic nature of the newlineCustomers It requires deep investigation and proper description of the newlinecustomer characteristics The task of analyzing the customer behavior is newlinesubdivided into two subtasks namely customer segmentation and target newlinecustomer analysis newlineCustomer segmentation gives a quantifiable way to analyze the newlinecustomer data and to distinguish the customers based on their purchase newlinebehavior It is typically done by applying a clustering algorithm to obtain a set newlineof segments The customer characteristics in each segment are analyzed by newlinemeans of target customer analysis The target customer analysis is achieved newlineby applying rule induction algorithm to generate rules or patterns to describe newlinethe behavior or characteristics of the customers in each segment There are newlinevarious clustering and rule induction algorithms serving different purposes newlineThe challenge involved in applying these data mining techniques to customer newlinerelated data is that these techniques should handle uncertainty in customer newlinebehavior newline newline