A statistical investigation on customer transactional manners

Data mining is a powerful new technique to help organizations mining the patterns andtrends in their customers data, then to drive improved customer relationships, and it is oneof well-known tools given to customer relationship management (CRM). This paperprovides a framework for segmenting customers in homogenous segments. Firstly, a cluster analysisis proposed to categorize the huge amount of data into several groups based on sixattributes: Deposit average, Rate of purchase with credit card, Bad cheque amount, number of bad cheque and the use of internet and telephone banking services. Then classifyingeach customer segment by Decision Tree (DT) under Analytical Hierarchy Process (AHP)technique based on expert people idea. Regression technique is used to implement the customer financial behavior model. The developed methodology has been implemented forprivate bank in Iran.

A statistical investigation on customer transactional manners Keywords:

Customer behavior , Key customer , Data mining , Clustering , Hierarchy process