Pattern recognition approach to customer analysis

This paper shows how to apply pattern recognition techniques to select products to sell to bank clients. The paper describes a method to select appropriate financial products to be offered to their existing clients. Cross selling is an important application of data mining for any company that offers its customers a wide range of products or services. Offering and selling more products to the existing customers makes them more valuable and more loyal. There are several approaches to building cross-sell models but this paper concentrates on one particular method that uses pattern recognition techniques. Experimental results are presented which demonstrate the feasibility of the proposed approach. The paper concludes with discussion and opportunities for future work