Accurate Marketing Of Credit Card Customers Based On AP Clustering Algorithm

With the improvement of informatization application in the world and the development of global economy,credit card has been widely popularized in daily economic life.The issuance of credit card has not only brought considerable profits to intermediary services such as Banks and special merchants,but also become a highlight of the profit contribution of the retail business of Banks.To keep the banking credit card business in sustainable way in the era of Internet finance is imperative to transform and update credit card marketing.The focus of this thesis is to dig out potential valuable customer groups from customer credit card transaction data.Accurate marketing information is pushed for customer groups with different characteristics,while stimulating the use frequency of credit card,improve the loyalty and satisfaction of credit card customers.This thesis took credit card customers' credit card data of a commercial bank as the research object,subdivided activeness of credit card customers in Affinity Propagation(AP)Clustering Algorithm way,then compared the results between fuzzy c-means(FCM)clustering algorithm and the k-means clustering algorithm analysis and verified the AP clustering algorithm's effectiveness in the credit card customer activeness segment,and established an effective customer segmentation model,finally designed the interactive interface,and provided quick query and marketing for customer segmentation.Through the research in this thesis,it is expected to make rapid and accurate marketing to bank credit card customers.This thesis aims to study cluster consumption data of credit card customers,to get characteristic model of customers according to the clustering results,and then to help specialized staff of bank credit card to identify the customer through the interface,to make precision marketing for each customer base with different needs,to attain customer's high experience and satisfaction in using credit cards,to reduce the risk of customers' loss.Major research work includes:?The consumption data of credit card was preprocessed,and the number of AP clustering algorithm was determined by the evaluation standard of clustering validity,and the clustering center was optimized by genetic algorithm to guarantee the accuracy of clustering results.? The AP clustering algorithm was used to subdivide the credit card holders,to construct valuable customer groups.The advantages and effectiveness of AP clustering algorithm were verified by comparing experiments of FCM clustering algorithm and k-means clustering algorithm.? Based on the optimal clustering results obtained by AP clustering,the detailed interactive prototype design of credit cardholder's activity was carried out,to realize precision marketing interactive system for credit card customers.